TestGuild Automation Podcast

Matt Wynne, co-creator of Cucumber and BDD practitioner, joins Joe for the first time in over a decade to talk about what two years inside a Silicon Valley AI startup taught him about the future of software testing.

Matt spent time at Mechanical Orchard working alongside experienced XP practitioners to modernize legacy COBOL mainframes using LLMs, and then spent a week with the team that coined the term "software factory," where the rule was simple: humans never write the code, never read the code.

In this episode, Matt breaks down what harness engineering actually means, why shared understanding is still the real bottleneck even in an agentic world, and how testers can use multiple LLMs to review AI-generated pull requests without reading every line.

He also gets honest about the grief that comes with realizing you can encode years of hard-won expertise into a Markdown file, and why that does not mean your skills are worthless.

If you are working in a brownfield codebase, wondering how to handle the flood of agentic PRs, or trying to figure out where testers fit in a world where agents write the code, this conversation is worth your time.

Find Matt at:

• mattwynne.net
• leansoftware.ai

Also check out his course: Build a Software Factory: Hands-off agentic coding for experienced engineers https://testgld.link/mattcourse

Amit Rawat is an agentic engineer who spent two decades in QA before shifting fully into building AI agents. He's the creator of PromptWright, a desktop tool that turns natural language prompts into automated Playwright browser tests, complete with screen recording, Gherkin scenario generation, and self-healing locators.

In this episode, Amit and Joe get into what it actually takes to work with AI agents at a high level, starting with why the planning phase matters more than the prompt itself. Amit breaks down his own workflow, brainstorming with AI, building a detailed plan in HTML before ever executing, and why curiosity and technical depth still matter even as AI gets more capable.

They also cover why Amit believes QA professionals, more than developers or DevOps engineers, are best positioned to thrive in the agentic era, how he tracks the ROI on his $200-a-month Claude subscription, the "Chief of Staff," "Chief Health Officer," and "Chief Financial Officer" AI agents he's built to help manage different aspects of his personal life, and how he uses a memory layer so those agents understand his preferences and become more useful over time.

If you're a tester, automation engineer, or QA leader trying to figure out where AI agents fit into your workflow and your career, this conversation is a practical look at what's already working today.

Most API testing stops at the happy path. The problem is that the bugs that actually hurt you in production are sitting in everything many testers skip, like the boundary values, the oversized payloads, the missing tokens, the security headers, the inputs that make no sense at all.

In this episode, Joe sits down with Liudas Jankauskas, who has spent almost twenty years breaking software and testing APIs since 2008.

Liudas demonstrates Rentgen, his free and open-source API testing tool, live on screen. You'll watch him take a single request from a real app, map it in seconds, and generate dozens of tests covering security, boundaries, performance, and load—all from one click.

You'll learn:

• How to discover APIs hiding under the hood of any application, even when there is zero documentation
• Why happy path testing leaves you exposed
• How to run a fast hygiene check before your real automation ever starts

Liudas also explains why Rentgen runs completely locally with no server and no data leaving your machine, making it safe for banking, healthcare, and other regulated environments.

Plus, he demonstrates the killer Copy Bug Report feature that drops a standards-based ticket straight into Jira or Trello.

• How to find and test undocumented internal APIs using the browser DevTools Network tab
• Why happy path-only testing misses the bugs that matter most
• How Rentgen turns one request into security, boundary, performance, and load tests automatically
• Where Rentgen fits in your workflow as a pre-automation hygiene layer—not a Postman replacement
• How to use it for regression by comparing results across environments
• The one piece of advice Liudas gives every tester to level up their API testing

Try Rentgen, free and open source, at Rentgen.io.

Connect with Liudas Jankauskas on LinkedIn: https://www.linkedin.com/in/liudas-jankauskas/

Your AI code review tools read the diff. They stare at your code. But they never actually run it. So the bugs that only show up at runtime, the broken user flows, the bad query plan, the duplicate submission, sail right past review and land in front of your customers.

In this episode, Joe Colantonio sits down with Evan Marshall, founder of Ito and a fifteen year engineer who spent five years in applied cryptography securing hundreds of millions of dollars for millions of people. Evan is taking that ship fast without breaking things discipline and pointing it straight at testing.

Ito is an agentic QA platform that builds and runs your actual app on every pull request, navigates it like a real user, exercises the frontend and backend as one system, and brings back real runtime evidence: video replays, logs, the exact lines responsible, and steps to reproduce, posted right in your PR.

• Why static code review misses the bugs that cause real production incidents
• How Ito spins up ephemeral environments and tests across UI, API, and database
• Why QA is not disappearing, it is leveling up into a manager and quality strategist role
• How to keep your test layer separate from your code generation so your signal stays honest
• The skills testers and engineers need as AI writes more of the code

If you are shipping AI generated code at high velocity and your QA cannot keep up, this one is for you.

Try Ito on your own code. Your first ten pull requests are reviewed free, no credit card required. Check it out at https://testgld.link/itoai now.

And as Joe always says, seeing is believing.

What happens to QA when AI is writing ten times more code than your team can test? That is the exact problem Ivan Barajas Vargas set out to solve with Amikoo, a purpose-built AI QA agent designed to help testers, SDETs, and even developers move faster without sacrificing coverage or quality.

Ivan is no stranger to AI in testing. Before generative AI became mainstream, he co-founded MuukTest, a test automation platform built on symbolic reasoning and expert systems. After six years and thousands of customer conversations, he went back to first principles to build Amikoo from scratch, this time with a harness of 12 specialized agents and 43 tools trained specifically for testing workflows.

In this episode, Ivan and Joe dig into the real-world gap between AI code generation and AI-powered testing, why the QA role is being elevated rather than replaced, how Amikoo uses Playwright and page object model patterns under the hood, and where human judgment still has to stay in the loop. Ivan also shares practical advice on what skills QA engineers should be building right now and which test scenarios should never be fully delegated to an agent.

If you are trying to figure out where testing fits in an agentic development world, this episode gives you a clear picture of what is possible today and what is coming next.

Visit https://testgld.link/amikoo to try the freemium account, and mention you heard this on TestGuild to unlock double the free usage.

Everyone is talking about AI replacing testers, writing tests, and transforming software quality. But what if we're asking the wrong question?

In this solo episode, Joe Colantonio shares a growing concern he's seen while traveling across the country for TestGuild IRL events: a decline in testing fundamentals at the exact moment AI hype is reaching a fever pitch.

Drawing insights from Carissa Véliz's book Prophecy: Prediction, Power, and the Fight for the Future, Wayne Roseberry's work on AI and meaning, and Tariq King's concept of Human Experience Testing, Joe explores why AI systems may be far less intelligent than many believe, and why human testers remain more important than ever.

You'll discover:

✅ Why large language models generate plausible answers without understanding truth

✅ The difference between prediction, correlation, and genuine understanding

✅ Why AI can test software but cannot experience software

✅ What "Everything is tested, but nothing is experienced" really means

✅ How AI hype may be distracting teams from critical testing fundamentals

✅ Why empathy, context, and human judgment are becoming competitive advantages for testers

Whether you're excited about AI, skeptical of it, or somewhere in between, this episode will challenge you to think more deeply about the future of testing and your role in it.

Resources Mentioned 📖 Prophecy: Prediction, Power, and the Fight for the Future by Carissa Véliz 📖 Work and presentations by Wayne Roseberry 🎓 Free course: thebullshitmachines.com 🎤 Learn more about TestGuild IRL events: TestGuild.com/irl

If you enjoy this episode, be sure to subscribe, leave a review, and share it with a fellow tester who's trying to navigate the AI era without losing sight of the fundamentals.

#SoftwareTesting #AI #QualityEngineering #TestAutomation #SoftwareQuality #HumanExperienceTesting #ArtificialIntelligence #TestGuild #QA #TechPodcast

AI coding tools promised to make development faster — and they delivered. But here's the problem nobody talks about enough: when you speed up coding, you don't eliminate the bottleneck in the SDLC. You just move it. And for most teams, it lands squarely in QA.

In this episode, Joe sits down with Vilhelm von Ehrenheim, Co-founder and Chief AI Officer of QA.tech, to dig into how agentic AI is reshaping software testing from the ground up. Vilhelm brings serious ML credibility, he helped build Motherbrain, one of the earliest production LLM systems in venture capital, and he's now applying that experience to one of the hardest problems in software delivery: testing at AI development velocity.

You'll learn how QA.tech's behavioral knowledge graph gives AI agents the context they need to actually understand your application, why validating user intent beats checking element identifiers every time, how autonomous agents can review PRs, reproduce bugs from Slack messages, and generate targeted tests without a single line of test code ,and what the tester's role actually looks like when agents do the heavy lifting.

If you're wondering whether your QA practice can survive the pace of AI-driven development, this one's required listening.

🔗 Book a demo now: https://testgld.link/qatechdemo

What happens when AI agents can not only write mobile app code, but also validate their own work automatically?

In this episode, I sit down with Maestro Co-founder and CEO Leland Takamine to explore one of the biggest shifts happening in software testing right now: agentic mobile testing.

Leland shares how his team went from solving mobile performance testing challenges to building one of the fastest-growing mobile automation frameworks used by companies like Microsoft, Meta, Amazon, and DoorDash.

We dive deep into:

• How AI coding agents are changing mobile testing workflows
• What "closing the agentic feedback loop" actually means
• Why deterministic testing still matters in the age of AI
• How Maestro MCP lets AI agents validate mobile app changes automatically
• Why mobile test maintenance may finally become manageable
• The future role of testers as AI-generated code explodes

Leland also gives a live demo showing an AI agent building, validating, debugging, and generating a reusable mobile test completely autonomously.

If you care about AI testing, mobile automation, MCP servers, or the future of QA engineering, this episode will likely change how you think about testing workflows over the next few years.

Try it out now for yourself:

• Maestro Studio: https://testgld.link/mstudio
• Maestro MCP docs: https://testgld.link/maestromcp

AI-powered testing tools are exploding across software engineering teams… but so are the hidden costs.

In this episode, Joe sits down with Arthur Hicken to unpack the growing problem of runaway AI token usage, unexpected LLM billing, and the operational risks of deploying AI agents into testing and DevOps pipelines. Inspired by Arthur's article on the emerging "Token Tax," this conversation explores why many teams are underestimating the true cost of AI automation.

You'll learn:

• Why AI-generated testing can create unexpected scaling costs
• How runaway AI agents and infinite loops happen
• Real-world examples of massive AI billing surprises
• Why deterministic problems shouldn't always use LLMs
• The hidden risks of "vibe testing" and autonomous AI remediation
• How QA teams can monitor, test, and control token usage
• Why performance testing and service virtualization matter more than ever in AI systems
• Practical strategies to avoid expensive AI deployment mistakes

Whether you're a software tester, automation engineer, QA leader, or DevOps practitioner, this episode will help you think more strategically about AI testing before costs spiral out of control.

AI-powered testing tools promise faster automation and less maintenance, but most require teams to abandon their existing frameworks.

In this episode, we explore Alumnium, an opensource AI-native end-to-end testing solution created by Alex Rodionov, an engineer at Airbnb and a tech lead on the Selenium project.

Instead of replacing tools like Playwright or Selenium, Alumnium adds an AI layer on top, helping teams:

• Reduce test maintenance by removing brittle locators
• Build more resilient, self-healing tests
• Write less code while increasing coverage
• Run tests across web and mobile using intent-based steps

We also go beyond the hype and break down what actually matters for real teams:

• Why AI-driven tests can still become flaky
• The performance and cost tradeoffs of LLM-based execution
• What "context rot" is—and how it impacts long test runs
• How to safely introduce AI into existing test suites without rewriting everything

Check it out now: https://testguild.me/alumAI