Featured Content | Product Updates | Community Updates | BrowserStack Testimonials

Featured Content

How High-Performing QA Teams Are Adapting To AI

[EVENT] The real challenge with AI isn't getting a demo to work. It's building quality workflows that teams can trust in production. At the QA Leadership Summit, engineering leaders and directors share how they're integrating AI into testing, measuring its impact, and creating reliable systems that scale. Virtual event, July 22.
Register Now!

Back to Top

Launching BrowserStack Load Testing

[NEW LAUNCH] Green API metrics shouldn’t hide a freezing UI when traffic spikes. BrowserStack’s new Load Testing platform unifies backend load generation, real browser performance, and APM metrics into a single dashboard. Powered by AI agents that automatically correlate data points and pinpoint root causes, it drastically cuts performance debugging time with zero infrastructure overhead.
Try Now!

Back to Top

Product Updates

App Automate| Percy & App Percy| Website Scanner| Accessibility | App Accessibility| Design Accessibility | Test Management | Low Code Automation

App Automate

• Test sharding for XCTestPlans: Accelerate your iOS testing with new sharding support for XCTestPlans, allowing you to easily distribute and run tests in parallel.
• Expanded Maestro support: We've broadened our Maestro framework integration to include full support for Emulators and Simulators, giving you more flexibility in your mobile testing.
• Enhanced interactive sessions: Enjoy a smoother testing experience with improved stability and full landscape orientation support during your interactive sessions.

Back to Top | Product Updates

Percy & App Percy

• Pinpoint location comments: Reviewing visual changes is now easier with Figma-like location comments, allowing your team to leave targeted feedback at specific points directly within a comparison.
• AI-powered ignore regions: Save time on manual reviews with new AI suggestions that automatically recommend ignore-regions for repeating visual differences at the same coordinates across your builds.
• Robot Framework support: We've expanded our testing ecosystem with full support for the Robot Framework.
• Latest SDK support: Stay up to date with our newly added support for Storybook 10 and Cypress 15.10 across the Percy SDK suite.
  
  Back to Top | Product Updates

Website Scanner

• Role-Based Access Control (RBAC): We've enhanced enterprise readiness with granular RBAC, giving you precise control over team permissions and data access.
• Mobile accessibility scanning: You can now scan for accessibility issues on mobile web pages with our newly added Chrome for Android support.
• WCAG-compliant home page: We are continuously improving our own accessibility, proudly achieving full WCAG compliance for the Website Scanner Home Page.
• Scan Time ETAs: Get better visibility into your scan progress with real-time ETA displays during execution.
• Enhanced report summaries: Gain deeper insights with our upgraded Report Summary Tab, now featuring Issue Resolution Trends, Issue Trends, and powerful Compare Reports capabilities (available for all scanners except A11y).
• One-year trend scores: Monitor your long-term website health with new 12-month historical tracking for the Broken Link and Spell Check scanners.
• Historical Core Web Vitals: Our Performance Reports now feature historical tracking of Core Web Vitals for your specified URLs, helping you monitor speed and user experience over time.
  
  Back to Top | Product Updates

Accessibility

• Smarter issue triage: Cut down your triage time and boost productivity with smarter issue grouping, perception management, and bulk "Needs Review" capabilities.
• Role-Based Access Control: Secure your data with our newly enabled RBAC for users and service accounts, ensuring authorized access across your organization.
• Redesigned navigation: Easily discover our full product suite and powerful AI capabilities with a newly optimized, user-friendly left navigation layout.
• Custom Dashboards (GA): Now generally available, Custom Dashboards let you consolidate all your reports to seamlessly track accessibility health and trends in a single, centralized view.
  
  Back to Top | Product Updates

App Accessibility

• Accessibility Scoring: Quickly assess, benchmark, and track your app's overall accessibility health with our newly launched A11y score.
• Certificate injection: You can now seamlessly and securely test enterprise apps with our newly enabled certificate injection support.
• AppLivery integration: Easily test the accessibility of apps distributed via AppLivery directly within your standard A11y workflow.
• Comprehensive CSV exports: Your CSV exports can now include passed elements, giving you a complete picture of your accessibility testing results.
• Enhanced Flutter compatibility: Enjoy improved accuracy and compatibility when scanning Flutter apps, thanks to our latest rule engine enhancements.
• Dedicated device support: Run your App A11y automated tests seamlessly on dedicated automated devices.
• Maestro and TRA integration: We've added specific Maestro enhancements to ensure smooth compatibility with Test Reporting & Analytics (TRA).

Back to Top | Product Updates

Design Accessibility

• Expanded GRR support: We've rolled out Global Region Routing (GRR) support across the EU, US, IN, and AU regions to ensure data compliance and performance.
• Universal Component Scanner: You can now scan any design component with our significantly expanded Component Scanner capabilities.
• Enterprise-grade RBAC: Manage your team securely with our new Role-Based Access Control, tailored to meet the strict security requirements of enterprise workflows.

Back to Top | Product Updates

Test Management

• Exploratory testing: You can now conduct flexible, unscripted exploratory testing directly within the platform.
• Xray Server support: Seamlessly integrate your workflows with our new Xray Server support for test cases.
• Enhanced Rich Text Editor: Document your Test Plans and Test Runs more effectively with our upgraded editor, now featuring robust table formatting, image support, and secure migration integrity.
• Interactive pie charts: Gain deeper visibility into your test coverage with interactive, drill-down pie charts in the Test Plan Detailed View.
• Smarter tag searching: Find the tests you need instantly with improved case-insensitive and partial matching for tag searches.
• Enhanced traceability reports: We've added a dedicated dataset column to our Requirement Traceability reports to give you a more comprehensive view of your coverage.
• Granular RBAC: Maintain strict control over your test management data with highly customizable Role-Based Access Control.
  
  Back to Top | Product Updates

Low Code Automation

• Enhanced performance and stability: Enjoy a faster, more reliable low-code automation experience with our latest under-the-hood stability improvements.
• API variable management: You can now programmatically set variable values via API calls directly within your Low Code/No Code (LCNC) workflows.
• Dynamic drag and drop: Build tests faster and more intuitively with our newly implemented dynamic drag-and-drop functionality.
• Global timeout settings: Gain more control over your test executions by easily configuring global timeout settings across your projects.
• Expanded device support: Test across a broader range of hardware with new devices added to our Low Code Automation platform.
• Unified status updates: Keep your team aligned with synchronized automation status updates between Low Code/No Code and Test Management.
  
  Back to Top | Product Updates

Community Updates

[Case study] TUI automates compliance and shifts left with BrowserStack Accessibility suite

Back to Top | Product Updates

BrowserStack Testimonials

Back to Top | Product Updates

We're delighted to announce that BrowserStack has been recognized with the 2026 TrustRadius Top Rated Award. This marks our fourth consecutive win since 2023!

With a trScore of 8.4 out of 10 and more than 260 verified reviews, BrowserStack continues to top the leaderboards across five key testing categories: 

1. Accessibility Testing 
2. Automation Testing 
3. Cross-Browser Testing 
4. Functional Testing 
5. Test Management

Supporting every stage of modern software testing

These categories represent the foundations of modern software quality. As development teams work to ship faster while meeting ever-higher expectations for user experience, robust testing across accessibility, functionality, compatibility, automation, and test workflows has become evermore important. We're proud that teams around the world continue to choose BrowserStack as their trusted testing platform, helping them build, test, and deliver exceptional digital experiences with confidence.

Advancing our vision for software quality

At BrowserStack, our vision is to be the testing platform for the internet. Every day, we help power the digital economy by enabling teams to build quality, bug-free software at speed; the very software that powers businesses, connects communities, and shapes how people live and work. Recognition across these critical testing categories reinforces our commitment to helping teams deliver great software experiences, no matter what they're building.

Powered by customer feedback

Across hundreds of reviews, customers consistently highlight BrowserStack's ease of use, extensive real-device coverage, reliability, and ability to help teams accelerate software delivery without compromising quality. 

Thank you to our customers

To everyone who took the time to share their experiences and feedback on TrustRadius: thank you. Your trust, support, and partnership continue to inspire us to raise the bar for software testing and help teams everywhere build better software, faster.

On International Women in Engineering Day, the global tech community comes together to celebrate the problem-solvers, builders, and innovators shaping our digital world.

This year, under the theme #EngineeringIntelligence, we are shining a spotlight on a powerful shift already happening across our teams: how women in tech are embedding artificial intelligence into their daily workflows as a core multiplier.

There is a common misconception that AI is here to replace engineers. The reality we see every day at BrowserStack is different. AI is an amplifier of human ingenuity. It handles the cognitive heavy lifting of parsing data and automating repetitive workflows, freeing engineers to focus on what humans do best: deep critical thinking, creative problem-solving, and strategic decision-making.

As you read through the stories below, you’ll see how our engineers are not just treating AI as a passive assistant. They use it as a technical challenger to stress-test their ideas and logic, rapidly synthesize complex data, and remove administrative overhead. Strikingly, that same mindset often carries into their personal lives—applying these tools to everything from home construction to family life.

Here is how the women of BrowserStack are driving innovation forward:

Snehal Thakkar – Senior Director of Products

“

#EngineeringIntelligence to me is using intelligence to make sharper engineering-product decisions, and building that same execution intelligence into the product.

My favorite AI use case right now as a product leader is using it as a strategy sparring partner grounded in real data.

The hardest part of the AI shift is discerning what’s actually true and figuring out where to place our bets. AI has compressed that strategic loop for me from weeks to a single afternoon.

I use it day to day to pressure-test strategies — turning a rough thesis into a defensible, evidence-backed point of view, with counter-arguments surfaced rather than hidden. It helps ground decisions in data, moving from a sharp question to an answer the same afternoon instead of the next sprint. It also keeps me current, synthesising what practitioners are genuinely learning about AI in the software lifecycle, so strategy reflects reality, not hype. It also helps turn research into artifacts — diagrams, one-pagers, and prioritization work ready to be put in front of leadership.

Outside of work, I use AI to buy back time—minimizing the hours spent planning and researching so I can focus on making memories with my family.

AI has made judgement faster and better-evidenced. And that is exactly what #EngineeringIntelligence means to me: using intelligence to make sharper engineering-product decisions, and building that same execution intelligence into the product.

Foram Vora – FullStack Engineering Manager

“

AI doesn't make engineering management easy, it buys back attention from busywork so I can spend it on what actually needs a human. To every woman building in tech: the judgment is the part that's yours. The tools just give you more room to use it.

This International Women In Engineering Day, the theme #EngineeringIntelligence made me reflect on how my relationship with AI has evolved. Most conversations about AI in engineering focus on writing code faster. For me, the real shift has been somewhere else entirely.

I'm an engineering manager leading 18 engineers across multiple projects at BrowserStack; two of them critical, parallel migration programs. On any given day regression cutoffs, blocked developers, shifting milestones and leadership updates, all compete for the same few hours. AI is what helps me hold all of it in view at once.

Instead of drowning in context-switching, I use it to keep a bird's-eye view across the work: surfacing regression readiness before every cutoff, flagging which developers are blocked or drifting, tracking milestone progress across the migrations, and turning scattered signals across Jira, Slack and our dashboards into clear decisions and crisp status updates.

It doesn't make the work easy — nothing does. What it does is buy back my attention from the busywork, so I can spend it on the parts that actually need a human: coaching my team, unblocking people, and making the harder calls.

That's what Engineering Intelligence means to me — not replacing engineering judgment, but clearing enough noise that there's room to use it well.

To every woman building in tech: the judgment is the part that's yours. The tools just give you more room to use it.

Rishika Sood – Frontend Engineering Manager

“

Whether it's navigating architecture discussions at work or structuring meal plans at home, AI serves as a powerful enabler that frees up time and cognitive headspace, amplifying creativity and productivity.

My favourite AI use case is using it as a thought partner that helps me navigate the many hats I wear as an engineering manager and beyond.

At work, I rely on AI for everything from architecture discussions and communication to accelerating workflows and reducing repetitive effort. Seeing how AI can free up engineers to focus on higher-value problems is something I find incredibly exciting.

In my personal life, I enjoy using AI to bring more structure to everyday routines. One simple but surprisingly impactful example has been building recurring reminders and meal-planning workflows for my son's tiffin. It takes away the constant mental overhead of figuring out "what's for tomorrow?" and helps me stay organized.

What I appreciate most about AI is that it gives me back time and headspace. Whether it's enabling teams to move faster or helping me manage daily life more efficiently, I see AI as a powerful enabler—one that amplifies human creativity and productivity.

Prity Ukey – Senior QA Engineering Manager

“

From cutting defect analysis timelines from days to hours to acting as a brilliant brainstorming and planning thought partner, AI is the ultimate multiplier when paired with deep domain expertise.

As a Quality Engineering Manager, AI is deeply embedded in my day-to-day work to help my team move faster, make better decisions, and focus on what matters most—shipping high-quality products.

It has drastically condensed our timelines—reducing production defect analysis from days to mere hours and cutting test automation efforts down to a half-day.

Beyond the metrics, it’s a brilliant thought partner that I use throughout the day. It helps in brainstorming test scenarios, outlining project plans, structuring stakeholder communication and more.

I also apply this structured optimization outside of work; during a recent home construction project, I used AI to analyze material quality, estimate costs, and build negotiation strategies with vendors.

While AI has become everyone's new friend, I believe in using it thoughtfully and purposefully. The more experience you have, the more powerful AI becomes as a strategic partner. It isn't about replacing expertise—it's about amplifying it.

Snehi Jain – QA Engineering Manager

“

Whether bridging architectural knowledge gaps with Copilot and Claude, or researching evidence-based parenting techniques, AI serves as an invaluable instant-context companion.

As an engineer, one of the most exciting changes I've witnessed is how AI has transformed the way we learn, connect ideas, and solve problems.

My favorite aspect of using AI is its ability to provide instant context and accelerate understanding.

Whether I'm onboarding to a new architecture, exploring an unfamiliar technology, or connecting complex concepts across systems, AI helps bridge knowledge gaps and enables me to contribute more effectively and confidently. Tools like GitHub Copilot and Claude have helped immensely in this regard.

As a working mother, AI has also become incredibly valuable beyond the workplace. It serves as a learning companion in my parenting journey, helping me better understand my son, discover insightful books and videos, and explore evidence-based techniques that support his growth and development. More importantly, it helps me reflect on how I can grow alongside him—balancing the demands of career and family while remaining intentional in both roles. My absolute favorite find in this journey has been the "The Power to Choose" book series for kids.

Hetvi Solanki – FullStack Software Engineer

“

I led a company-wide Developers Round Table, where we demystified AI and MCP integrations for teams to cut down manual effort by up to 85%. Making AI more accessible to other builders has been the most rewarding part of the journey.

My favourite AI use case isn't one tool — it's getting an entire team (and beyond) to actually use AI well.

As someone helping lead the AI adoption on the Test Management team while also being a Full-Stack Engineer, I get to weave AI across our day-to-day engineering, from writing frontend optimized code and tests to drastically reducing API turnaround times and speeding up how we resolve issues. The part I'm proudest of is leading a company-wide Developers Round Table (DRT), where I walked engineers through the AI tools and MCPs (Model Context Protocol integrations) we can plug into our workflows — and showed, with real examples, how we can cut down the tedious parts of our process so we can focus on the actual engineering.

A lot of this instinct comes from the hackathon circuit. Winning 16 hackathons over the years, including BrowserStack's own AI-First Initiative Hackathon, has instilled in me a "build fast, use every tool available" mindset. And that is exactly what I try to bring to the team. During BrowserStack's hackathon, we built and released an end-to-end workflow for production tickets in 24 hours, that handled initial triage, team assignment, and even provided a fix in the form of a PR without a single manual intervention, cutting down human effort by 70–85%.

Demystifying AI for other developers has easily been the most rewarding part of this journey.

Shubhi Singh – Software Engineer in Test

“

If I could tell one woman thinking about engineering one thing, it'd be this: AI isn't here to replace the curiosity that got you into this field. It just clears the noise so you can spend more time on the parts you actually fell in love with.

When I started out in QA automation, the part of the job I dreaded wasn't writing tests — it was everything around it. Our Live and App Live regression work is scattered across several repositories: the test framework in one, the app code in another, shared helpers in a third (and my sanity somewhere in a fourth that I never found).

Before I could automate anything, I'd spend hours hopping between repos like a detective with too many tabs open, trying to reconstruct how a feature actually behaved versus how it claimed to behave. The real engineering was buried under a mountain of "wait, where does this function even live?"

My favorite AI use case rescued me from that mountain. And honestly, calling it a "pair-programmer" undersells it — it's more like an autonomous teammate who has read the entire codebase and, unlike me, remembers all of it. I hand it a whole task and it does the unglamorous legwork — reading across all those repos, tracing how a feature works, pulling in the right helpers — and comes back with working automation for me to review. What's left for me is the fun part that actually needs a human: deciding what's worth testing, hunting edge cases, and the judgment calls. The work feels lighter — and a lot more like thinking again (and a lot less like crying into a stack trace).

It doesn't stop at work, either. At home I've leaned into the same idea — small automations to offload the mental clutter. My favorite is using AI to plan my week: I dump in my chaotic to-do list, deadlines, and the seven reminders I set and ignore, and it hands back a calm, prioritized plan — like a personal assistant who is more organized than I am (and judges me only slightly).

So, if I could tell one woman thinking about engineering one thing, it'd be this: AI isn't here to replace the curiosity that got you into this field. It just clears the noise — and the 47 browser tabs — so you can spend more time on the parts you actually fell in love with.

Puja Jagani – OpenSource Lead & Dev Advocate

“

My favorite way to leverage AI is as a challenger rather than an assistant. It’s incredible how the same technology that challenges my engineering thinking and helps uncover blind spots can also ease the daily mental load of parenting.

My favorite way to leverage AI is actually as a challenger rather than an assistant.

I frequently ask AI to critique my thinking, disagree with my proposals, find gaps in my code, push back against my technical reasoning, surface missed edge cases and more. It forces me to think more critically and refine my engineering and problem solving approach.

This knack for optimization translates into my personal life, too. I recently returned from maternity leave and my younger daughter has just started solids. I use AI to map out meal plans and double-check that we are covering the right nutritional needs through the week.

It’s incredible how the same technology that sharpens my engineering logic can also seamlessly make everyday parenting decisions a little easier.

Nithya Mani – Senior Lead Customer Engineer

“

By using GitHub Copilot for rapid technical demos and NotebookLM for on-the-go learning, AI helps me work smarter. It frees up my energy for what matters most: human judgment and collaboration. Beyond work, it has also unlocked new creative outlets in ways I never imagined.

AI has become an essential multiplier across both my professional and personal life.

At work, I leverage GitHub Copilot to quickly build sample repositories for customer demos, event showcases, and debugging complex technical issues. I’ve built custom Gemini Gems for account intelligence gathering, Mutual Success Plan creation, research, and other recurring workflows. I also use AI regularly to analyze customer feedback, identify trends, and accelerate problem-solving.

NotebookLM has become my go-to companion for continuous learning. The Audio Overview feature helps me convert lengthy documents, reports, and research material into bite-sized audio summaries that I can consume while walking, commuting, or doing household chores.

I'm currently experimenting with Claude workflows to automate day-to-day tasks and improve productivity as well. I am still early in the journey, but excited about the possibilities.

Beyond productivity, AI has given me a fun and creative outlet too. I used ChatGPT, DALL·E, and Canva to create personalized invites, games, and creative content for a friend's wedding. It turned out to be a huge hit and reminded me how creative AI can be.

What I love most is that AI hasn't replaced any part of my job—it has helped me become more efficient, learn faster, and spend more time on work that truly requires human judgment, creativity, and deep collaboration.

Aayushi Singh – Senior Customer Engineer

“

By handling the administrative heavy lifting across complex testing frameworks, automation tools, and customer environments—summarizing meetings, surfacing historical context, refining communications, and creating clear documentation—AI frees me to focus on what matters most: delivering better outcomes for our users.

One of my favourite AI use cases is how it helps me be a better problem-solver and communicator in my role as a Customer Engineer.

Navigating a vast landscape of testing frameworks, automation tools and customer environments means my days are packed with complex issues, log analysis, escalations, and cross-functional meetings. AI acts as a collaborative assistant that optimizes my entire workflow.

Managing back-to-back conversations while ensuring that key details, decisions, and action items are captured can be challenging. Post-meeting, I feed my raw notes into AI to instantly summarise issues, extract action items, identify blockers, suggest logical next troubleshooting steps, and draft polished customer follow-ups. It helps to refine customer communications, create documentation, summarize discussions, and even prioritize work effectively. It’s also incredibly effective for digesting historical contexts like long email threads, previous issues or support tickets, and account handovers. It helps me quickly understand the background, identify important details, and get up to speed faster.

What excites me most is how AI amplifies engineering thinking. By stripping away the administrative overhead of customer engineering, AI allows me to dedicate my energy to deep collaboration, problem-solving and driving better outcomes for our users.

Andrea Alvares – Automation Support Engineer

“

AI serves as both a troubleshooting partner and a technical mentor. AI doesn’t replace my technical judgment; it gives me the space to think deeper and solve problems faster.

In my role, AI serves as both a troubleshooting partner and a technical mentor.

When debugging complex customer issues, I treat it as a sounding board to explore hypotheses, validate my thought process, and uncover troubleshooting paths I might have missed. It helps me look at problems from different perspectives while still relying on my own technical judgment to reach a solution.

It is particularly powerful when dealing with massive log volumes, like Appium or device execution logs. Instead of manually scanning thousands of lines, I use Gemini Pro to rapidly isolate failure points, allowing me to focus entirely on root-cause investigation. I’ve even built custom Gemini Gems based on historical resolutions to fast-track solutioning. The gems help me quickly ground with past learnings, eliminate known possibilities, and speed up troubleshooting when similar issues arise.

Ultimately, AI doesn’t replace my technical judgment; it gives me the space to think deeper and solve problems faster.

(Responses may have been edited for clarity.)

#EngineeringIntelligence is about empowering brilliant engineers to build, test, and scale with unprecedented clarity and speed.

If you manage accessibility for a large-scale website, you know the sinking feeling. You run an automated scan across your website, hoping for a clear roadmap to compliance. Instead, the engine spits back a laundry list of over 10,000 reported issues.

Your development team spends days fixing genuine accessibility bugs, yet the overall issue count barely drops.

Why? Because traditional accessibility engines report the exact same component error—like a missing aria-label in your global site header—as a unique issue on every single page. One broken header on a 500-page site generates 500 unique alerts.

This creates massive issue bloat. While these violations are technically real, the sheer volume feels like a system error. It creates the perception of false positives, makes reports feel incredibly noisy.

Today, BrowserStack is changing the narrative. We’ve completely overhauled the BrowserStack Accessibility Testing reporting experience to help you filter the noise, verify report reliability, and cut your triage time by up to 60%.

Here is a deep dive into the new workflow.

1. Intelligent issue grouping to eliminate report bloat

Seeing 500 identical issues clutters your backlog, hides your team's actual progress, and makes it impossible to know how many unique problems actually need fixing.

We have built an advanced, intelligent grouping logic into the core of our Spectra Rule engine. BrowserStack now automatically consolidates identical or similar issues, both within a single page and across multiple pages. Your issue count will now reflect the true scope of your accessibility health, eliminating report bloat.

2. Confidence % view and slider to restore trust in reports

Navigating a high volume of accessibility data can make it difficult to distinguish between immediate high-impact fixes and more nuanced potential issues, often slowing down the path to remediation.

We are giving you complete transparency with a new Confidence % view that shows the exact accuracy of every reported issue.

Using the Confidence Slider, you can instantly filter out low-confidence noise and prioritize high-probability, critical violations first.

3. Guided bulk review to save ~60% in triage time

Manually reviewing and dismissing the exact same component error hundreds of individual times is a tedious, frustrating experience that typically requires deep WCAG expertise.

Leveraging our new grouping logic, we’ve built a split-screen Guided Bulk Review workflow that provides full visual DOM context.

Instead of reviewing pages individually, anyone on your team can confirm or dismiss hundreds of grouped issues instantly by answering simple, plain-English questions—no WCAG expertise required.

4. Frictionless triage to track your progress and boost your score

Triage often feels like an endless treadmill where clearing out issues takes so much time that your accessibility score rarely reflects your team's actual effort.

Because you can now filter noise and dismiss non-issues in bulk, you can rapidly clear your backlog with near-zero friction.

This streamlined workflow allows you to efficiently track your review progress and see tangible, dynamic improvements to your overall accessibility score.

Ready to cut through the noise and triage your accessibility issues faster than ever?

Log in to your dashboard to see the new UI, or Start a free trial to experience high-confidence, low-noise accessibility testing today.

Day 3 of Breakpoint 2026 was built for execution. After two days of keynotes, debates, and product launches, the final day gave engineers and QA practitioners the technical blueprints to act on what they had heard. Two intensive masterclasses ran across the morning. Fourteen lightning talks ran alongside them, ten minutes each, practitioner to practitioner, covering everything from building mobile use agents with open-source models to why AI still cannot replace the judgment that testing actually requires.

The Masterclasses

Master AI Testing: Join the Top 1%

This was the deep end. The session moved past basic ChatGPT experimentation into the mechanics of building and deploying autonomous test generation agents from scratch. Using LangChain and CrewAI as foundational frameworks, the masterclass walked through how to integrate custom agents into existing testing ecosystems like Selenium, Playwright, and Cypress, turning fragile test suites into self-healing architectures.

The most technically demanding section tackled AI's biggest practical challenge: non-determinism. Attendees left with working techniques for flexible assertions, RAG-based validation, and observability tooling to trace and debug every autonomous decision an agent makes.

Watch this masterclass.

Master QA's Next Shift: MCP-Powered Testing

The second masterclass addressed a problem most QA teams recognise: crucial context is scattered across Jira, Figma, GitHub, and CI/CD pipelines, and pulling it together for root-cause analysis wastes time that compounds across every sprint. The session showed how the Model Context Protocol (MCP) acts as a bridge across those fragmented tools, and walked attendees through building production-ready MCP workflows from scratch in VSCode with GitHub Copilot and BrowserStack. The practical output: faster debugging, automated test case generation from context, and a replicable workflow for reducing release risk. Watch it for a hands-on blueprint to MCP that goes beyond theory.

Watch this masterclass.

The Lightning Talks

4 Nuggets on Automating AI Evaluation

In non-deterministic systems, manual "vibe checks" don't scale. Elias Pardo shared four hard-earned lessons on turning a 4-hour evaluation process into a 30-minute automated pipeline, covering failure modes, tooling choices, and how to codify human judgment without losing what makes it useful.

Watch this session.

A Developer's Love Letter to QA

After 15 years on both sides of the bug report, Bob Fornal, Senior Solutions Developer at Leading EDJE, explored what makes the Dev/QA relationship both essential and exhausting. He unpacked the root causes of cross-team conflict and made the case for reframing QA feedback as a collaborative act rather than a verdict, with practical communication strategies for teams stuck in that cycle.

Watch this session.

NL to Mobile Actions: Building an Open-Source Mobile Use Agent with LangGraph, Qwen VL, and Appium

Ahmad Waqar, Lead Engineer at FYI.ai, built a mobile use agent that takes plain English instructions and executes them autonomously on real Android and iOS devices, with no proprietary LLMs and no API key required. He walked through the three-pillar architecture powering it: LangGraph for stateful orchestration, Qwen vision and language models served locally via Ollama, and Appium for actual device interaction, along with an honest look at how open-source vision models actually perform against proprietary alternatives.

Watch this session.

Stop Experimenting with AI, Start Shipping It!

Siddhant Wadhwani, SDET Manager and Head at Newfold Digital, started with one problem: slow test case creation. Five months later, that became ATLAS AI, a RAG-enabled engineering copilot used daily across testing, development, and product teams. He shared the architectural decisions that made it scalable, the two features that drove overnight adoption, and the mistakes that nearly derailed the project, all wrapped in a concrete "Start Small, Scale Fast" framework.

Watch this session.

Shift Left is Old. Are You Doing Shift Smart?

Kapil Jain, QA Automation Engineer at PayPay India, team tested early and still shipped critical defects to production. His diagnosis: shift-left without prioritisation just means more tests, not better ones. He introduced Shift Smart, a pipeline restructuring that moved from UI-heavy upfront automation to API-first validation and contract testing, then showed the defect leakage and deployment stability numbers before and after.

Watch this session.

Quality Is Not a Metric — It's a Business Signal

Boards fund confidence and risk control, not test coverage percentages. Soumen Deb, Director at LTIM, introduced a three-layer translation framework for converting raw engineering telemetry (flaky rates, escaped defects, test failures) into board-ready signals like Revenue at Risk and Customer Impact Probability, with a one-page executive snapshot template and a six-week pilot plan to get started on just two critical business flows.

Watch this session.

From Bug Hunters to Risk Managers

Modern QA's value is no longer measured by bugs found or test cases run. Saloni Tongia, Lead QA Engineer at InfoBeans, made the case for QA teams repositioning as risk advisors, with practical guidance on building risk awareness, strengthening stakeholder trust, and turning quality engineering into a function that shapes release decisions rather than just reports on them.

Watch this session.

Symbiosis Between the Community and the Individual

Mirza Sisic, QA Consultant at TestOps, made the case that the relationship between a tech community and its members runs both ways: neither thrives without the other doing the work. He shared how attending and organising tech meetups helped him move past social shyness and build lasting professional confidence, and why active participation delivers returns that passive membership never does.

Watch this session.

Smart Testing with Playwright Network Interception

Waiting on real backend responses and managing third-party dependencies makes test suites slow and unreliable. Vijayaraghavan Vashudevan, QA Specialist at Natwest Group, demonstrated how Playwright's network interception capabilities (filtering API responses, blocking unnecessary resources, mocking backends) create deterministic test environments that run faster and break less, without depending on backend availability.

Watch this session.

Building Automation Suites with AI, Avoiding Its Seduction

Olubukola Omotayo, Director of Software and Quality Assurance at Home Trumpeter, built four complete automation suites using AI coding agents, covering functional UI testing, accessibility, and security, and came away with a clear conclusion: AI is a skill amplifier, not a replacement for judgment. The session covered where AI genuinely accelerates automation, where it cannot assess risk based on dynamic context, and why its greatest danger is what it convinces you to stop questioning.

Watch this session.

The "Verifier" Pattern: Why Autonomous Agents Fail in QA and How to Fix Them

Most LLM agents writing tests fail because they are treated as black-box creators, and "mostly correct" is not good enough in high-stakes engineering. Shankha Subhra Bagchi, Vice President at Blackrock, introduced the Verifier Pattern: a decoupled architecture where a generator agent is governed by a separate, constraint-based verification layer. The practical outcome is a self-healing loop that moves QA from human-in-the-loop to human-on-the-loop, cutting AI debt and maintenance overhead in the process.

Watch this session.

Multi-Agent Swarms for Autonomous Edge-Case Discovery

Standard automation tests what you already know to look for. Ishan Katoch demonstrated what happens when you turn a multi-agent framework loose on a production UI with the sole objective of breaking it. He walked through a QA Swarm architecture with three specialised agent roles: an Explorer to map DOM state, an Attacker to inject chaotic inputs, and an Observer to log exactly when and how the system fails, and showed how to deploy the whole thing against your own staging environments.

Watch this session.

From Manual to AI-Augmented

The shift from manual test design to AI-augmented coverage is already happening inside real engineering teams. Ricardo Cristalli walked through practical strategies for implementing AI augmentation without losing test quality, covering what the change actually means for how QA teams think about coverage, risk, and day-to-day test design decisions.

Watch this session.

Build Your Own Testing Playground: How QAs Can Grow Automation Skills by Creating Simple Apps

QA engineers who want to practice automation consistently hit the same wall: no stable system to test against. Bruno Nascimento de Figueiredo, QA Automation Engineer at Mestre QA, answer was to build one. A local API becomes a training ground for contract validation, error handling, and backend automation. A small frontend becomes a safe environment for UI selectors and end-to-end flows. He walked through a real student example and gave attendees a clear starting point they could act on the same day.

Watch this session.

All Sessions Are Free to Watch

Day 3 delivered what the first two days set up: the technical detail to actually act on the vision. Two masterclasses gave engineers working architectures for production-grade AI testing. Fourteen practitioners shared what they built, what they measured, and where they drew the line on letting AI do the thinking for them.

That was Breakpoint 2026. Three days, across AI-native quality engineering, enterprise-grade proof, and practitioner blueprints. All sessions are free for registered attendees.

Read the Day 1 recap here. | Read the Day 2 recap here.

Day 2 of Breakpoint 2026 moved from concept to proof. Where Day 1 made the argument for AI-native quality engineering, Day 2 ran the benchmarks. Speakers from KPMG, UST, Capco, Microsoft, and CGI spent the day testing the uncomfortable questions — whether traditional test engineering still has a future, what AI actually changes in day-to-day QA work, and how teams sustain quality when building speed is no longer the bottleneck.

The Proof is the Product: Securing and Scaling Enterprise AI

Every enterprise AI claim eventually meets three tests: can your security team approve it, can it scale to your environment, and can it do what basic LLMs cannot. Four BrowserStack product leaders ran live benchmarks and architecture deep-dives to answer all three, moving the Day 2 conversation away from vision and into hard evidence. Watch it if your team is evaluating enterprise AI testing and needs proof beyond the pitch.

Watch this session.

To Infinity and Beyond! — The Death of Test Engineering

Keith Klain, Director of Quality Engineering at KPMG, made a case that a lot of people think privately but rarely say out loud: traditional test engineering has failed to deliver on its ROI promise, and AI is accelerating the reckoning. He argued that self-healing systems and generative AI platforms are finally addressing the stagnant software quality and rising headcount costs that have been accumulating for decades. It is a harsh session, and deliberately so. Watch it if you have sat through testing ROI conversations that felt disconnected from reality.

Watch this session.

QA in 2030: What Changes, What Stays, What Disappears

Mobin Thomas, Global Head of Quality Engineering at UST, brought two decades of enterprise experience to cut through the AI hype. His session focused on what is actually changing inside real enterprise QA teams, where AI is delivering value and where it is not, and what mindset shifts practitioners need to keep pace. Less prediction, more practitioner's lens. Watch it for a ground-level read on where the profession is heading, from someone who has seen multiple cycles of transformation from the inside.

Watch this session.

Designing Work That Works: A Job Crafting Approach to Testing

Ashley Hunsberger applied organizational psychology research to a question most QA professionals have felt but not named: how do you find meaning in a role that keeps changing around you? She introduced the concept of job crafting, intentionally shaping your role around your strengths, values, and core motivators like autonomy, variety, and impact. The session challenged the idea that growth means working harder, making the case that soft skills like learning from failure and embracing discomfort are what actually unlock potential in a rapidly shifting field.

Watch this session.

Centaur Mode: A QA Engineer's Honest AI Playbook

Brittany Stewart, Senior QA Specialist and Engineer at QualityWorks Consulting Group, shared what integrating AI into real QA work actually looked like: trial, error, and a lot of honest reflection about where it helped and where it got in the way. The model she landed on is centaur mode — the human stays firmly in the loop, AI amplifies what you already bring. She walked through her actual workflow changes, the metrics she tracked, and what you can start applying immediately. Watch it for a no-hype account of human-AI collaboration in practice.

Watch this session.

Resilient Quality in the Age of AI: A Debate on Code, Craft, and Commoditization

Two distinct philosophies collided in this panel. On one side: technical architecture, self-healing frameworks, scalable test engineering. On the other: testing craft, risk analysis, irreducible human judgment. Rahul Parwal Test Specialist at ifm, and Naveen Khunteta, Test Automation Expert at AutomationLabs debated the uncomfortable questions — whether self-healing suites mask deeper problems, whether AI can ever decide if software is good enough, and where a tester's real value lives when machines handle execution.

Watch this session.

Leveraging AI Code Assistants to Build, Test and Automate Tests for APIs

Julio de Lima, QA Manager at Capco, did a live build. He wrote an API, designed tests for it, automated them, and pushed them through a pipeline — using AI code assistants at every step. The session bridged the gap between API and UI testing, showing how to use GenAI responsibly to accelerate test design, automate functional and non-functional suites, and deploy them directly into pipelines. Watch it if you have been circling API test automation and want to see how it actually fits together with AI tooling in practice.

Watch this session.

Ditching the Bloat: Meet the Next-Generation Requestly API Client

Heavy, bloated API testing tools with mandatory logins and cloud lock-ins have been a long-standing frustration. This BrowserStack Unpacked session introduced the next-generation Requestly API Client along with a first look at an AI-enabled API Test Authoring Agent built to automate complex test generation directly inside your workflow. Watch it for the official product demo and to see what a faster, privacy-first API testing experience looks like.

Watch this session.

The Accessibility Quadrants

Ady Stokes, an Accessibility Advocate has spent years focused on WCAG compliance before recognising it was only one piece of the accessibility picture. His session introduced the Accessibility Quadrants, a visual heuristic that shifts teams from checking compliance boxes toward a more complete view of inclusion — one that accounts for readability, inclusive language, and overall usability alongside technical conformance with frameworks like WCAG and the European Accessibility Act. Watch it if your team treats accessibility as a compliance exercise and wants a more useful frame.

Watch this session.

Behavior-Driven Context Engineering

Behavior-Driven Development was built to close the gap between product and testing. Andrew Knight's session asked what happens when your newest collaborator is an AI coding agent. He explored a practical framework for applying BDD to AI-assisted development, showing how structured context files and explicit rules guide agents through the full development lifecycle, from discovery and Gherkin scenario formulation through to agentic code generation and iterative repair. Watch it if you use BDD and want to understand how to extend it into AI-native workflows.

Watch this session.

From MVP to Production-Grade: Guardrails in the Age of AI

Building products is no longer the bottleneck. With AI agents, ideas can become working MVPs in days. The real challenge has moved to sustaining quality and preventing degradation as systems evolve. Pramod Yadav shared his experience building an MVP with AI agents and, critically, using those same agentic capabilities to introduce intelligent quality gates. The session focused on designing systems that balance speed with reliability when code is increasingly generated and modified by autonomous systems.

Watch this session.

Exploring with Code: When Testing Stops Splitting in Two

The traditional divide between manual testing and automation has long been a structural problem. Maaret Pyhäjärvi's session reimagined exploratory testing for the DevOps era, making the case that actionable results are the only testing coverage that truly matters. She explored how to capture testing insights programmatically, expand the tester's role from bug reporter to active debugger, and use AI as external imagination to shift focus from bug hunting toward exploring intent, value, and user experience. Watch it if your team still runs manual and automated testing as two separate tracks.

Watch this session.

How Hard Can It Be — To Make Sense

Error messages, warnings, and validation alerts are often the most informative things a system produces — and the most ignored. Lena Nyström showed how treating these messages as a trail of clues, from crisp frontend validations to convoluted database notifications, reveals which part of the tech stack generated them and why. That knowledge, she argued, is one of the most underused tools in a tester's kit. Watch it for a practical, experienced take on reading what a system is actually trying to communicate.

Watch this session.

Prompt Injection Is the New SQL Injection: Rethinking Testing in the AI Era

Traditional software fails when code breaks. AI systems fail when reasoning breaks. Gaurav Khurana, Senior Test Consultant at Microsoft, introduced AI Red Teaming from a tester's perspective — not as a cybersecurity exercise but as an evolution of exploratory testing for intelligent systems. The session covered how AI applications fail through prompt injection, hallucinations, context poisoning, and false confidence, connecting those failure modes to familiar testing concepts like boundary testing, fuzzing, and risk analysis.

Watch this session.

All Sessions Are Free to Watch

Day 2 covered the hard questions: what traditional test engineering still owes the profession, how teams build AI into daily QA without losing judgment, and what it takes to sustain quality when building is no longer slow. Speakers brought proof rather than vision — live builds, real metrics, and frameworks that hold up outside demo environments.

Day 3 shifts to hands-on implementation: agent building, MCP-powered testing workflows, and lightning talks from practitioners sharing what they tried, what they measured, and what actually changed. Read the Day 3 recap here.

Breakpoint 2026 opened on 12 May 2026 with an argument the QA community has been building toward for years: the speed-versus-quality trade-off in software testing is no longer a real constraint. Across the first day's keynotes, product launches, and fireside chats, engineering leaders from NVIDIA, Mastercard, Teladoc Health, Microsoft, Motorola Solutions, and Genpact made the same case from different angles, when AI handles execution, human judgment becomes the work, not the obstacle to it.

Quality or Speed: Why You No Longer Have to Choose

For years, the practical choice was: ship fast and accept the risk, or slow down and hold the standard. This keynote argues that AI-native quality engineering makes that choice obsolete. When AI handles test execution, engineering teams stay focused on judgment rather than throughput.

Watch this session.

The End of Code as IP: Redefining Competitive Moats in the AI Era

With AI generating code from specs, the code itself is no longer your primary competitive moat. What cannot be replicated are your specs, your PRDs, and your test suites — they encode domain knowledge and product understanding that models cannot infer. Watch this if you want to rethink where engineering value actually sits in an AI-driven org.

Watch this session.

The Pioneer's Playbook: Non-Determinism and the AI-Native QA Era

Jason Huggins separates what is genuinely changing in AI-era testing from what is hype. He focuses specifically on where AI is producing real, observable change in API, performance, and security testing, and directly challenges the decade-old shift-left mantra. Worth watching for any team that wants ground-level clarity rather than vendor talking points.

Watch this session.

Building an AI-Native Quality Operating System: From Manual QA to Intelligent Quality Engineering

Mary Jinugu walked through what it actually looks like to replace a traditional QA model rather than layer AI on top of it. Her team moved to Copilot-driven automation, intent-based workflows, and agentic QA systems. Watch it for a step-by-step account of what changes, what breaks, and what you have to rebuild.

Watch this session.

Testing AI Agents: From Confident Answers to Trustable Outcomes

AI agents fail in ways that look completely fine until they don't. Deepshikha and Anamika introduced a three-layer framework for testing agents across reasoning, actions, and end-to-end outcomes, covering prompt injection, data leakage, and how to set quality gates for systems without deterministic outputs. Watch it if you are testing or planning to deploy AI agents in production.

Watch this session.

Scaling Your QA Career: From Individual Contributor to VP

Richard Hand's trajectory from individual contributor to Assistant VP gave this session a practical spine. He talked about what career growth in QA actually requires beyond technical skill: influencing cross-functional teams, tying testing decisions to business outcomes, and why senior leaders sometimes create the exact problems they ask QA to solve. Honest, specific, and useful at any career stage.

Watch this session.

Guarding the Brand: Leveraging AI to Catch 'Invisible' UI Regressions at Scale

A 0.5px padding error passes every functional test but erodes brand consistency across thousands of pages. Harit Narke showed how Mastercard moved from manual UI reviews to an automated Visual Review Agent in the CI/CD pipeline, cutting a multi-day sign-off process to a 15-minute automated report. Watch it if visual quality at scale is something your team is currently managing manually.

Watch this session.

Building Gender-Expansive Teams in Tech

Jenny Bramble, Sophia McKeever, and Jenna Charlton moved well past surface-level DEI language to discuss what actually makes environments safe and inclusive for trans, non-binary, and cis-woman professionals in tech. The panel covered pronoun usage in team settings, holding space in male-dominated sprints, and why psychological safety directly affects the quality of engineering work.

Watch this session.

Have Some Cake With Your Frosting: Testing Both the UI and API Layers

Most testing attention drifts toward the UI, but the business logic and the hardest bugs live in the APIs underneath. Hilary Weaver covered why API testing is not just a back-end concern, how HTTP response codes fit into a testing strategy, and how to translate UI test cases into API calls. Watch it if your team is over-indexed on UI coverage and wants a practical path to better test coverage.

Watch this session.

What 80% Accuracy Actually Costs: A Production Claude Code and MCP Pipeline

Getting to 80% accuracy with AI code generation tools is achievable. The drag from the remaining 20% is what teams aren't accounting for. Suneet Malhotra walked through a production-grade MCP (Model Context Protocol) pipeline built specifically to close that gap, including how to address AI debt and build guardrails that hold in high-stakes enterprise environments.

Watch this session.

Two Major Product Launches: Test Companion and BrowserStack Evals

Breakpoint 2026 was also the stage for two significant BrowserStack product announcements.

• Test Companion is an IDE-embedded AI assistant for generating and debugging test scripts, designed to keep engineers in their workflow rather than context-switching to external tools.
• BrowserStack Evals addresses one of the defining problems of AI-native testing: how do you measure the output of a system that does not produce deterministic results? BrowserStack Evals is built specifically to validate non-deterministic AI outputs at scale, which makes it a genuinely new category of tool rather than an extension of existing frameworks.

All Sessions Are Free to Watch On Demand

Day 1 covered a lot of ground: two product launches, the case against code as IP, a new framework for testing AI agents, and practitioners from Mastercard, NVIDIA, and Microsoft showing what the shift to AI-native quality engineering looks like in production.

Day 2 shifts from vision to proof: enterprise AI testing benchmarked against real security and scale requirements, run live.

Featured Content | Product Updates | Community Updates | BrowserStack Testimonials

Featured Content

Breakpoint 2026: 50,000+ Attendees, 75+ Experts — Now On Demand

[EVENT RECAP] One question ran through every session at Breakpoint 2026: what does AI-driven testing look like when it's working? Selenium founder Jason Huggins, KPMG's Keith Klain, and 70+ speakers answered from real teams, not slide decks.
Watch everything on demand!

Back to Top

The State of AI in Software Testing 2026 is Here!

[NEW REPORT] 61% of teams now use AI across testing workflows, but the gap between high performers and the rest is widening. See what the data says about budgets, ROI, and where AI is delivering the most value.
Download the report!

Back to Top

Product Updates

App Automate| Percy & App Percy| Website Scanner| Accessibility | App Accessibility| Test Reporting & Analytics | Test Management | Low Code Automation

App Automate

• Maestro framework support expanded: You can now run tests for React Native, Webview, and Flutter applications directly on App Automate.

Back to Top | Product Updates

Percy & App Percy

• Faster CI failure debugging: You can now easily pinpoint issues with surfaced logs that highlight errors and suggest clear paths to resolve them.
• Smarter comparison logic: You can now save time with new SHA pair matching, which automatically skips redundant visual comparisons when the source and target SHAs remain unchanged.
• Enhanced collaborative feedback: You can now communicate more effectively with your team using new, Figma-style comments directly within your visual reviews.
  
  Back to Top | Product Updates

Website Scanner

• Broader authentication support: You can now test more complex authentication flows with newly enabled support for local and session storage.
• Reduced scan failures: You can now avoid common setup issues and minimize failed scans with an intuitive new cookie configuration nudge.
• Upgraded Accessibility standards: You can now experience full WCAG 2.2 AA compliance across all Web A11Y product touchpoints.
  
  Back to Top | Product Updates

Accessibility

• Partial page scanning launched: You can now target and test specific ownership areas with new partial page scanning for Accessibility in LCA and automated tests.
• Safari browser support (Alpha): You can now expand your macOS coverage with Safari support across the Website Scanner, Live Accessibility Toolkit, and Automated A11Y scans.
• Switch Tab support enabled: You can now seamlessly switch tabs during Accessibility automated testing to cover more complex, multi-tab user workflows.
  
  Back to Top | Product Updates

App Accessibility

• Compare Reports launched: You can now easily track Accessibility regressions by comparing successive runs of the same build side-by-side.
• Customized Accessibility reporting: You can now create clean, customized test reports by excluding specific accessibility rulesets on a test-level basis.
• Android screen reader enhancements: You can now experience improved full-screen screen reader (SR) automation scanning on Android with new linear navigation enhancements.
• Role-based access control (RBAC): You can now secure your organization's data and govern user permissions with newly launched role-based access control.

Back to Top | Product Updates

Test Reporting & Analytics

• Build deletion enabled: You can now delete builds alongside archiving them directly from the New Dashboard for better build management.
• Support for buildIdentifier: You can now automatically merge Automate or App Automate Non-SDK builds with matching values for a more unified reporting experience.
• Session IDs added to testRuns API: You can now access Session IDs directly through the API to seamlessly build custom reporting on top of your BrowserStack data.
• Emulator and Simulator support: You can now track, analyze, and manage App Automate builds running on emulators and simulators directly within TRA.

Back to Top | Product Updates

Test Management

• Pie chart drilldown enabled: You can now click into pie charts within the Test Plan Detailed View to gain deeper, more granular insights into your test data.
• Storage management improvements: You can now manage your testing resources more efficiently with newly upgraded storage management capabilities.
• Folder-level test case search: You can now quickly locate specific test cases by searching directly within individual folders.
• Attachment previews launched: You can now instantly preview file attachments directly in your workspace without needing to download them first.
  
  Back to Top | Product Updates

Low Code Automation

• Browser version support: You can now select specific browser versions for your test runs to ensure precise cross-browser compatibility.
• Smart tab handling: You can now seamlessly navigate and manage multiple browser tabs during your automated test workflows.
  
  Back to Top | Product Updates

Community Updates

[Case study] TUI automates compliance and shifts left with BrowserStack Accessibility suite

Back to Top | Product Updates

BrowserStack Testimonials

Back to Top | Product Updates

Most testing practitioners pick a side: open-source community work or enterprise platform building. Andrew Knight, Senior Director of Product Management at Cycle Labs, has spent his career refusing to.

He builds enterprise automation platforms by day and stays one of the most visible open-source advocates in the testing community in the hours around it. The two roles inform each other in ways that show up clearly in his work, and they shape how he thinks about where AI is taking software testing next. He's joining us at Breakpoint 2026 to talk about exactly that.

You've built a fantastic personal brand alongside your work at Cycle Labs. How does your passion for open-source advocacy influence the enterprise platforms you build?

Open source projects are a fantastic foundation upon which to build, even for enterprise platforms. For example, at Cycle Labs, our web browser automation is based on Selenium WebDriver. I also take inspiration from the cool things that open source projects do. Recently, I've been enamored with Playwright's AI tooling, namely its CLI skills and MCP server, and I'm looking to build similar features into the Cycle platform.

Can you give us a sneak peek into your Breakpoint session?

My Breakpoint session is entitled "Behavior-Driven Context Engineering." Although BDD has been around for almost two decades, its principles are more pertinent now than ever.

Through context engineering, AI tooling now enables us to generate both code and tests from well-written specs and rules. When done with a thoughtful human in the loop, the results can be outstanding. I want everyone to know that, with AI superpowers, if they can describe it, then they can do it!

What's a belief you had about enterprise software quality a year ago that has completely changed in the face of new AI testing tools?

A year ago, I believed that achieving high-quality test coverage for complex enterprise platforms would always be slow, manual, and heavily dependent on deep institutional knowledge.

AI testing tools have changed that by making context a first-class input. Through intentional context engineering, teams can encode their understanding of systems, workflows, and edge cases in a way that AI can immediately apply to generate meaningful test plans, cases, and even executable scripts.

It shifts the bottleneck from writing tests to curating and refining the context that drives them.

You focus heavily on strategic frameworks to empower global communities. What is the most common structural mistake enterprise teams make when integrating complex infrastructure?

The most common mistake is treating complex infrastructure as a tooling problem instead of a systems problem. Teams often layer new platforms on top of existing ones without aligning ownership, communication paths, and decision boundaries.

As complexity grows, the lack of clear interfaces leads to bottlenecks and inconsistent outcomes. The fix isn't more tooling; it's designing intentional structures that make dependencies, responsibilities, and feedback loops explicit.

When you finally close the laptop, what does a perfect day of switching off look like?

I like to play with my French bulldogs, work on my classic cars, and maybe play some video games.

Andrew is one of several practitioners bringing technical depth and real-world perspective to Breakpoint 2026. He'll be joined by Keith Klain, Jason Huggins, Marc Boroditsky, Avinash Ahuja, Lena Nyström, Brittany Stewart , Pramod Yadav, and more. The lineup is built for practitioners who want sessions grounded in actual work. Register here to catch his session.

Our Breakpoint 2026 speaker spotlight series continues with Julio de Lima, QA Manager at Capco. By day, Julio leads QA at one of the world's largest tech consultancies. Outside business hours, he runs a YouTube channel teaching software testing to the Brazilian community and delivers tutorials at major industry events. He's been studying AI in testing since 2018, when he started his master's degree on the subject, well before it became the topic everyone wanted to talk about.

At Breakpoint 2026, he's running a live build session in which he'll write an API from scratch, design tests for it, automate them, and push them through a pipeline using AI code assistants like Copilot.

You're a QA Manager at Capco by day and an educator with a YouTube channel teaching testers in Brazil. How do you balance both?

My teaching activities take place outside of business hours, which leaves me free to do both without one interfering with the other. Working at Capco motivates me every day because there's always a new challenge and this dynamic environment where I'm always challenging myself and learning more. Regarding my role as a teacher and content creator, I'd say that's the fuel that keeps me connected to the community.

For me, teaching is more than just being in charge of something; it's about seeing my students grow and become relevant to the market, and thus achieve their personal and professional goals. I learn, I share, someone learns, uses what they've learned, and grows. Is there anything more wonderful than that?

Can you give us a sneak peek into your Breakpoint session?

Sure, I've been delivering a three-and-a-half-hour tutorial at some of the world's leading software testing events, and I decided to bring a shorter version of it to Breakpoint so our community can see and understand how AI can contribute to software testing activities in practice.

In this talk, we learn how to build an API from scratch, create automated tests, and run them in the pipeline. Participants will see this entire process and leave full of ideas on how to build and test software with AI; it will be fantastic.

What's a belief you had about AI in testing a year ago that you've since changed your mind on?

My perception of AI has remained the same since 2018, when I started my master's degree in AI and Software Testing: It has helped me to be much more productive as a professional tester. The world we live in today is a golden age, where we can leverage the machine's capabilities to help us accelerate our work as QAs.

It's not about asking and accepting, but rather about asking technically and reviewing what has been brought to you with attention and care. This attention, care, and technical skill are what differentiate a professional tester who uses AI responsibly from the rest.

Most testers know API testing matters, but when it comes to getting started, where do people get stuck?

I believe it's the absence of a graphical interface and the greater possibility of data variation (given the freedom to fill payloads with any information). The point where I started to understand how to deal with this was when I realized that API testing is divided into 3 parts: testing the business rules, testing the API interface (whether it meets what its specification describes), and non-functional factors (such as backward compatibility, performance, reliability, interoperability, etc.).

If we think of it as a component that connects to other components, and define these other components that call it as personas, the modeling of your tests will be much simpler. Remember also that APIs are part of a more technical universe, so understanding more about databases, software architecture, queues, messaging, logs, etc. will make you more comfortable when working on their tests.

What do you like to do when you're away from the screen?

I enjoy longboarding, bodyweight exercises, running, playing my instruments (the flute is my current instrument, before it was the cello), and spending time with my wife and daughter at Disney parks.

Julio is one of several practitioners bringing hands-on perspective to Breakpoint 2026. He'll be joined by Keith Klain (Director of Quality Engineering, KPMG), Jason Huggins (Founder, Selenium and Vibium), Marc Boroditsky (Technology Leader, Nebius), Avinash Ahuja (Manager Solution Architect, NVIDIA), Lena Nyström (CEO, Test Scouts AB), Brittany Stewart (Senior QA Specialist, QualityWorks Consulting Group), Pramod Yadav (Test and Automation Specialist, Power Tester), and more. It's a lineup built for testers who want to leave with something they can use on Monday. Register here to catch Julio's session.

Our Breakpoint 2026 speaker spotlight series continues with Keith Klain, Director of Quality Engineering at KPMG. Keith brings something the testing industry doesn't get enough of these days: 25+ years of pattern recognition. He's seen which "next big things" actually delivered, which ones quietly disappeared, and what it costs to bet on the wrong one.

Most of his career has been inside the world's largest financial institutions, where the cost of a bad release isn't just downtime — it's regulatory exposure. He writes about the industry at qualityremarks.com, tracking the people and ideas actually shaping how testing gets done versus how vendors say it should. At Breakpoint 2026, he's presenting "The Death of Test Engineering", a session that, in his own words, comes from "20+ years of pent-up frustration."

After 25 years leading QE at major financial institutions, what does the work actually look like, and what's kept it interesting?

The work looks exactly the same as anywhere else that builds and deploys software, but I think one of the primary differences is the operational structure and oversight that comes with enterprise technology. When you get into certain types of financial services organisations, the context is just different than other industries because the exposure to regulatory and market moving consequences are higher.

As well, there is a corporate structure you have to deal with when despite not being able to conduct business without technology, it's not your primary product. So a lot of times you have to navigate business structures and operations that are counterintuitive to running a tech team, but are vital to building and supporting the business.

It can be very stressful at times, but I think that's one of the things that keeps it interesting and feeling you've very connected to global events, politics, etc. keep it fresh and current and I'd probably be bored working in any other industry!

Can you give us a sneak peek into your Breakpoint session? What's the one thing you want everyone to walk away with?

Aside from unloading on the test automation industrial complex for 20+ years of pent up frustration, my talk is a challenge to one of the biggest assumptions in our industry: that test automation means more/better testing and higher quality. I'll also lay out how the old model of test engineering built on script production, regression sprawl, and maintenance-heavy automation has lost relevance in an AI-driven world.

Hopefully you'll walk away with some ideas on how we got here and a clear message: don't build your future around writing more tests, build it around understanding risk, failure, and system behaviour. The people who stay relevant won't be the ones producing the most test scripts. They'll be the ones who can ask better questions, model what matters, detect problems earlier, and explain why systems fail.

What's a belief you had about AI in testing a year ago that you've since changed your mind on?

What has changed for me is the full realization of just how powerful the cognitive surrender to AI has been from people I used to respect in this business. I think I underestimated the power of anthropomorphism with GenAI and that it's pretty much treated like magic. It has created a mania and FOMO that is nothing like I've seen before, so my belief that systems thinking, risk management, and oversight could counteract that has been completely changed.

What hasn't changed it that most of my pessimistic views about AI in testing have pretty much been validated, specifically how bad actors (consultants and vendors) in our industry are cynically just hyping AI as some evolutionary next step for quality. If you've been around this business long enough, you see the familiar patterns from people who clearly don't understand testing, don't care about testers, or have to live with the consequences of their quality decisions.

To be clear, a tester's JOB is to be skeptical about things in order to get an objective view about progress, quality, and most importantly risk. So it's getting pretty boring listening to the usual suspects drone on about how questioning the value of AI in testing and looking at risk is just being negative or acting like some kind of Luddite.

What are some of the best tech-focused newsletters, blogs, or podcasts you'd recommend?

I'm a pretty heavy consumer of everything testing industry related and am very active on LinkedIn and industry events, but my go to sources for keeping up on the industry is Software Testing Weekly, the Software Testing Round Up by Alan Richardson, the Engineering Quality podcast, and some Google/Reddit alerts I've set up.

To keep up with AI and the endless stream of news stories, I keep my inputs limited to trusted sources to focus my attention, and those are: Dagmar Monett, Emily M. Bender, Timnit Gebru, Charity Majors, Karen Hao, Fiona Charles, Michael Bolton and Maaike Brinkhof.

I've also outlined most of who and what I listen to on a regular basis that has influenced my thinking about my work on my Resources page, and more specifically for AI here.

Outside of the tech world, what's a hobby or passion you keep coming back to?

Well, I spend a lot of time at work so I take every chance I get to hang out with my family and travel. I hiked the South Downs in Surrey last year by myself and it was amazing and I'm hoping to do another one this year in Cornwall.

Keith is one of several seasoned voices bringing hard-earned perspective to Breakpoint 2026. He'll be joined by Jason Huggins (Founder, Selenium and Vibium), Marc Boroditsky (Technology Leader, Nebius), Avinash Ahuja (Manager Solution Architect, NVIDIA), Lena Nyström (CEO, Test Scouts AB), Brittany Stewart (Senior QA Specialist, QualityWorks Consulting Group), Julio de Lima (QA Manager, Capco), Pramod Yadav (Test and Automation Specialist, Power Tester), and more. It's a lineup built for practitioners who want signal over noise. Register here to catch his session.

Our Breakpoint 2026 speaker spotlight series continues with Brittany Stewart, Senior QA Specialist and Engineer at QualityWorks Consulting Group. Brittany brings a perspective you don't often see in this space: she came into testing through design, and that crossover shapes everything about how she works.

She's spoken at TestBash World, SauceCon, and Agile Testing Days 2025 (where her session ranked in the top 10 rated talks), and has documented what AI-augmented workflows actually look like in practice on her blog at brittanystewart.tech. At Breakpoint 2026, she's presenting "Centaur Mode: A QA Engineer's Honest AI Playbook."

You came into QA from a creative and design background, which is not the usual path. How has that shaped the way you approach testing?

Design trained me to care about the person on the other side of the screen first. So when I look at a test, I'm not just asking "does this pass?" I'm asking "does this actually protect the user's experience?" That shift matters more now that AI is in the workflow.

Right now I'm deep in test automation for a large enterprise application, building out an AI-augmented workflow that covers test scenario generation, smart test healing, and scoring the quality of AI output. I'm not trying to automate everything. I'm trying to make the repetitive work faster so human judgment stays sharp where it actually matters.

Can you give us a sneak peek into your Breakpoint session? What's the one thing you want every attendee to walk away with?

My session is called "Centaur Mode: A QA Engineer's Honest AI Playbook." I built this framework because I kept watching QA folks fall into one of two traps with AI. Either they refused to touch it and got left behind, or they handed everything over and started shipping whatever the model generated. Neither works. What I want people to understand is that collaborating with AI is a skill on its own, and it's one QA engineers can actually get good at.

Centaur Mode is the middle path. Picture a centaur. Human on top, horse on the bottom, working together, but the human is steering. That's the mental model I want people to leave with. AI can be your advantage instead of your replacement, but only if you stay in control of the judgment calls. You don't need to be a prompt engineer to pull this off. You need a repeatable process, clear handoffs, and the discipline to review every output.

What's a belief you had about AI in testing a year ago that you've since changed your mind on?

A year ago I thought the goal of AI in testing was to get it to do more of the work. More generation, more coverage, more output. I spent a lot of energy there, and honestly, that was the easy part.

What I didn't see coming was the harder part. Knowing when to trust the output and when not to. And the mental capacity it takes to properly evaluate what AI gives you back. That's a different job than what QA has traditionally done, and I don't think we talk about it enough.

AI does get better the more you use it. The tests start matching your patterns, the structure looks right, but every now and then it will inject something that's not actually relevant to what you asked for. You have to catch it. Generation is fast and cheap for everyone now. Judgment and evaluation are the real job, and that takes mind power, experience, and knowledge. As humans in this loop, we have to protect that part.

"Centaur Mode" is a pretty memorable framework. Where does the human absolutely have to stay in control, and where is it actually okay to let AI take the wheel?

The human has to own intent and acceptance. What are we testing, why does it matter, and is this output actually correct and meeting business needs? Those three questions belong to you. Hand them off and you're not doing QA anymore — you're just shipping whatever the model felt like generating that day.

Where I let AI take the wheel is the in-between work. Drafting a first pass of scenarios from a ticket. Setting up the starter structure for feature files and step definitions. Self-healing a selector that drifted. Summarizing a failed run and explaining root cause. The repetitive, pattern-heavy stuff where a fast draft beats a blank page. But every handoff back to me goes through a quality check. That's the whole point of Centaur Mode.

You're based in Jamaica, you run a design studio, and you work in one of the fastest-moving corners of tech. What does "switching off" actually look like for you?

It looks like being present with my husband and my son. My son is in Grade 3 and he does not care about my test suite, but he's very into video games, crafts, and tech. So we play games together and explore building our own games. We also take trips and spend time in nature whenever we can. "Touch some grass," as my son would say. That's grounding in the best way.

I still make time for the design and marketing work I love too. Switching off for me isn't really about doing nothing. It's about building something that matters outside of work.

Brittany is one of several practitioners bringing grounded, real-world perspectives to Breakpoint 2026.

She'll be joined by Jason Huggins (Founder, Selenium and Vibium), Marc Boroditsky (Technology Leader, Nebius), Avinash Ahuja (Manager Solution Architect, NVIDIA), Keith Klain (Director of Quality Engineering, KPMG), Lena Nyström (CEO, Test Scouts AB), and more. It's a lineup built for practitioners who want signal over noise. Register here to catch her session.

Featured Content | Product Updates | Community Updates | BrowserStack Testimonials

Featured Content

Breakpoint 2026: The Playbook for AI-Driven Testing Is Being Written Right Now

[EVENT] Most QA teams are navigating AI-driven testing without a clear roadmap. Breakpoint 2026 is where that changes — Selenium founder Jason Huggins, KPMG's Keith Klain, Ashley Hunsberger, and over 50 speakers share what's actually working across real teams. Live sessions, honest Q&A, hands-on workshops. Join us from wherever you are, 12–14 May 2026.
View the agenda and register!

Back to Top

The State of AI in Software Testing 2026 is Here!

[NEW REPORT] 61% of teams now use AI across testing workflows, but the gap between high performers and the rest is widening. See what the data says about budgets, ROI, and where AI is delivering the most value.
Download the report!

Back to Top

Product Updates

App Automate| Percy & App Percy| Website Scanner| Accessibility | App Accessibility| Accessibility Design Toolkit|Accessibility Devtools| Test Reporting & Analytics | Test Management | Testing Toolkit

App Automate

• Maestro gets major upgrades: You can now run tests on the latest v2.x and split your suite into shards for faster parallel execution across real iOS and Android devices.

Percy & App Percy

• Browser upgrades for Percy Web: We've upgraded support for Chrome, Edge, Safari, and Firefox, ensuring your visual testing environments stay perfectly up to date.
• Unified search experience: Finding exactly what you need is now faster and easier with our enhanced, unified search functionality across the Percy dashboard.
  
  Back to Top | Product Updates

Website Scanner

• Dynamic URL authentication: You can now interact with and click elements directly within the Website Scanner to seamlessly authenticate and run reports on gated flows.
• Biannual automated scans: Set it and forget it with our new six-month recurring scan support, giving you reliable, long-term automated monitoring with proper database persistence.
  
  Back to Top | Product Updates

Accessibility

• Part of Page Scanning: Now, you can focus your scans on the specific components or sections you own - across Test automation and Low Code Automation
• Safari support for Live Toolkit (Alpha): We are expanding our testing coverage for macOS users with our new Safari browser support, currently available in Alpha.
• WCAG 2.2 AA compliance: You can test with confidence knowing our platform now fully meets the latest WCAG 2.2 AA accessibility standards as well.
  
  Back to Top | Product Updates

App Accessibility

• Compare Reports: Easily track and squash accessibility regressions by comparing successive runs of the same build side-by-side.
• Customizable rulesets: Generate cleaner, highly tailored reports by excluding specific accessibility rulesets on a test-level basis during automated runs.
• Enhanced Screen Reader Automation: You can now automatically scan the entire screen, even elements beyond the visible viewport, to catch deep screen reader violations.
• Role-Based Access Control (RBAC): Admins now have granular control over user permissions and actions within their organization, keeping your test data secure and preventing unauthorized access.

Back to Top | Product Updates

Accessibility Design Toolkit

• AI Contrast Checking for Gradients & Images: Our AI agent now automatically detects color contrast for text on gradients and images in your Figma files.
• Figma plugin exports: Easily keep your team in the loop by exporting scan results directly from the Figma plugin to your Design toolkit dashboard, simplifying how you share accessibility statuses with stakeholders.
  
  Back to Top | Product Updates

Accessibility Devtools

• AI-Powered Custom Component Mapping: You can now use our VS Code Chat Participant to automatically map your internal Design System components to accessibility rules.
• Expanded linting support: We've broadened our developer toolkit by introducing dedicated linting support for the UIKit framework, HTML, and Angular, helping you catch issues earlier in your code.
  
  Back to Top | Product Updates

Test Reporting & Analytics

• Pattern-based failure analysis: Save time on manual triage. You can now define custom regex-based "pattern rules" on test error messages and stack traces to automatically tag failures with the correct category.
• App Automate Detox support: App Automate users can now seamlessly view their Detox testing sessions directly within the New Dashboard, eliminating the need to toggle back to the old interface.
  Back to Top | Product Updates

Test Management

• Shared preconditions support: Streamline your test creation by easily setting up and reusing shared preconditions across your test suite.
• Inline test case editing: Make updates on the fly with new inline editing capabilities directly within the Test Case Details slideover.
  
  Back to Top | Product Updates

Testing Toolkit

• Multi-language test cases: Break down global barriers by writing and managing your test cases in any language.
• Standalone Atlassian generation: Our test case generator now works directly seamlessly on Atlassian pages without requiring any extra integration setup.
• AI-powered API chaining: Supercharge your Common Workflow Automation by using AI to effortlessly chain complex API calls together.
• Enhanced Storybook support: You can now leverage advanced Storybook features with new support for MDX, Autodocs, and Shadow DOM, giving you even more flexibility in how you build and visually test your modern UI components.
  
  Back to Top | Product Updates

Community Updates

[Case study] TUI automates compliance and shifts left with BrowserStack Accessibility suite

Back to Top | Product Updates

BrowserStack Testimonials

Back to Top | Product Updates

Breakpoint 2026 is back from 12–15 May, fully virtual and free, and it’s shaping up to be one of those editions where you actually want to show up for the sessions, not just bookmark them.

We’re kicking off our speaker spotlight series with someone who’s seen the industry evolve from the inside: Lena Nyström, CEO of Test Scouts AB. She’s been building software since 1999, starting out as a developer before moving into testing, which gives her a perspective that’s both technical and deeply practical.

At Breakpoint this year, she’s diving into something most teams underestimate: system messages. The kind you see every day — errors, warnings, popups — that seem simple on the surface but are surprisingly hard to get right. And more importantly, what those messages reveal about how systems are actually built.

You've been building software since 1999 and moved from QA to CEO of Test Scouts AB. How has your definition of "quality" evolved?

I actually started as a developer and did that for a decade before moving into testing. And to be honest, I don't think it's changed much.

I've always been very interested in quality from requirement creation to working in production — that's just how I was schooled. It likely shaped how I've worked with software in all of my roles and is also the foundation to why I always struggled to understand why other people saw testing as something limiting and small.

For me, quality starts with the idea and ends when software is decommissioned. The code is just a tiny part :-)

Can you give us a sneak peek into your Breakpoint session? What's the core message you want to deliver?

I will talk about something I find absolutely fascinating: how incredibly difficult it is to write clear, understandable messages of different kinds. Error messages, information messages, warnings, popups.

And it LOOKS so simple, and is so funny when it goes wrong. Not to mention the fascinating differences we can see historically and how much you can learn about architecture and systems by studying them.

Most teams treat system messages as an afterthought. But they’re often the first place users experience failure — and the only clue teams have when something breaks.

What's a belief about sustainable technological infrastructure you've shifted your perspective on?

I am terribly sorry but again, I am not sure much HAS shifted. Maybe five to ten years ago I believed I was terribly outdated and that I moved too far away from deep tech to be relevant — but the more I see and hear, the clearer it is to me that while everything changes, everything stays the same.

Old ideas resurface and the core just... stays relevant.

After navigating the tech industry for over two decades, what's your approach to long-term burnout prevention and work-life balance?

I hope I have an answer to this before I retire but to be honest — I am terrible at this. I give way too much of myself and work is too much of my identity.

Lena is one of several leaders joining us at Breakpoint 2026 this May. If there’s a common thread across the sessions this year, it’s this: the way we think about quality is changing — but the fundamentals still matter more than we’d like to admit. Sometimes, it’s the smallest things — like a system message — that tell you the most.

Breakpoint is back for 2026, and this year the conversation is a bit more honest.

Most teams already have AI somewhere in their testing workflows. The hard part now isn't adoption. It's the messy middle: integration problems, test suites that were supposed to get faster but somehow got more complicated, and the quiet uncertainty about what all of this means for your team long-term.

Breakpoint 2026 (12–15 May 2026, virtual, free) is built around exactly that moment: after the pilot, before everything clicks.

What is Breakpoint?

Breakpoint is BrowserStack's annual virtual conference for the global software testing community. It started in 2020 and has run every couple of years since, growing to 20,000+ attendees by 2025. It's always been free.

The format hasn't changed much across editions: practitioner-led sessions, no fluff, real workflows from people who've actually shipped the thing they're talking about.

This year's theme is Testing, Reimagined by Intelligent AI. Because most teams are somewhere between "we've adopted AI" and "we've figured out AI," and that gap is where the interesting work is happening.

Some of the people on stage (more to come)

The full speaker lineup is still coming together, but here are a few of the confirmed sessions:

• Julio de Lima (QA Manager, Capco) is doing a live build. He'll write an API, design tests for it, automate them, and push them through a pipeline, all using AI code assistants like Copilot. The value isn't the tools, it's watching someone who actually does this daily show you how it fits together. If you've been circling API test automation without knowing where to start, this one's for you.
• Brittany Stewart (Senior QA Specialist, QualityWorks Consulting Group) is presenting what she calls "Centaur Mode." She's logged real numbers from her own practice: 75% less time creating test scenarios after working AI into her workflow. But the session isn't about the win. It's about everything that went wrong first, and the model she landed on where the human stays firmly in the loop.
• Keith Klain (Director of Quality Engineering, KPMG) is asking out loud something a lot of people think privately: has test engineering actually delivered what it promised? His session is called "The Death of Test Engineering" and it's not a gentle take. If you've ever sat through a testing ROI conversation that felt disconnected from reality, this will feel like fresh air.
• Lena Nyström, CEO of Test Scouts AB, is presenting on how to read what your system is actually telling you when something breaks. Error messages, validation failures, alerts. She's been building software since 1999 and the session is as practical as that experience suggests.
• Pramod Yadav (Test and Automation Specialist, Power Tester) went from idea to shipped SaaS MVP in a few days using AI agents, and then used those same agents to build quality gates so the thing didn't break the moment he added a feature. It's a first-person account of agentic testing in the wild, not a walkthrough of what the tools claim to do.

Day 3 is a full run of lightning talks: 10-minute sessions from practitioners sharing what they tried, what they measured, and what actually changed. More sessions and speakers are being added, so check the event page for the latest.

Is this for you?

Whether you're a QA engineer looking for technical depth, an engineering manager thinking about team structure in an AI-driven world, or a developer who cares about what happens after code ships, Breakpoint has content built for you.

Sessions span beginner-accessible to advanced practitioner level, and there are dedicated tracks for both hands-on practitioners and people in leadership roles.

Want to understand the bigger picture before May? Check out the State of AI in Software Testing 2026 report, which surveyed 1,000 software professionals on how teams are adopting AI in testing, where they're seeing ROI, and what's coming next.