Jay McCure

Last week, during a customer discussion, one question kept coming up again and again: "How do you ensure your AI is telling the truth?" Not faster… not smarter… but truthful. Because in the world of AI, hallucinations are not just technical errors—they’re business risks. A confident but incorrect response can impact decisions, customer trust, and even compliance. At KiwiQA Services, we’ve been working closely with organisations to move beyond just building AI… to validating AI in production. From detecting hallucinations, measuring factual accuracy, to reducing inconsistencies across responses—this is where real AI assurance begins. We’ve captured our approach, frameworks, and insights in this blog: https://lnkd.in/g4zH-Fnz If you’re deploying LLMs or planning to -this is a must-read. Because the future of AI isn’t just intelligent… it’s reliable. If you're looking for AI hallucination testing, get in touch with us. #AI #LLM #AIQuality #AIHallucination #AIValidation #KiwiQA #AITesting

27

1 Comment

𝘞𝘩𝘺 𝘠𝘖𝘜𝘙 Automation 𝘧𝘳𝘢𝘮𝘦𝘸𝘰𝘳𝘬 𝘸𝘪𝘭𝘭 𝘍𝘈𝘐𝘓 𝘪𝘯 6 𝘮𝘰𝘯𝘵𝘩𝘴 (𝘢𝘯𝘥 𝘩𝘰𝘸 𝘵𝘰 𝘧𝘪𝘹 𝘪𝘵) I just audited a $50M company's automation framework. Result? 73% of their tests were failing randomly. Their CTO asked me one question: "How did we go from 10 passing tests to complete chaos?" The Brutal Truth: 85% of Selenium Projects Die the Same Death Month 1: "Our automation is amazing!" Month 6: "Why does everything break when we deploy?" Here's what kills every framework (and the fix): 1️⃣ The "Everything in One Folder" Disaster ❌ Death pattern: UI, API, utils all mixed together ✅ Fix: Dedicated packages → UI, API, POJO, services separated Reality check: Good teams onboard new devs in 2 hours, not 2 weeks. 2️⃣ The "Hardcoded Hell" Problem ❌ Death pattern: URLs, data, timeouts scattered everywhere ✅ Fix: Environment property files + externalized test data Game changer: Switch DEV→QA→PROD with one command. 3️⃣ The "No POJO = No Scale" Trap ❌ Death pattern: Raw JSON strings, manual API payloads ✅ Fix: Request/Response POJOs + schema validation Impact: API tests become 10x more maintainable. 4️⃣ The "Debug Nightmare" Issue ❌ Death pattern: "Test failed" with zero context ✅ Fix: Extent Reports + screenshots + API logs Truth: Debug time drops from 2 hours → 5 minutes The Framework That Actually Scales I've built a production-ready structure that includes: 🏗️ Proper separation of UI/API/POJO layers 🔧 External configurations for all environments 📊 Rich reporting with screenshots & metrics 🚀 CI/CD ready with Docker & Jenkins support 🎯 BDD structure that business teams understand The Bottom Line: Stop building "quick automation scripts." Start building software systems that scale. Your framework should work at test #10 AND test #1000. Want the complete folder structure? 👇 Comment "𝑭𝑹𝑨𝑴𝑬𝑾𝑶𝑹𝑲" and I'll send it to your inbox! Found this helpful? Share with someone struggling with flaky tests! 🚀 -x-x- Full Stack QA & Automation Framework Course with Clearing SDET Coding Rounds: https://lnkd.in/g7tn6Uif #japneetsachdeva

156

10 Comments

DDD Melbourne speaker spotlight: Callum Whyte Failure will happen. In software, that’s not a risk—it’s a guarantee. Code breaks, servers crash, power goes out – yet building software that is designed to fail gracefully isn't always front-of-mind. Resilient systems don’t avoid failure… they’re designed to fail gracefully. Callum Whyte will walk us through how to design and build real-world systems that keep working when things go wrong. Expect a practical, engineering-first deep dive into: – Handling failure at every layer – Redundancy, circuit breakers & automated failover – Architectural tradeoffs: latency, consistency & degraded modes – Building reliability into team processes and mindset – Using health checks, OpenTelemetry & stress testing tools to stay ahead of issues This session is packed with tools, patterns, and techniques you can bring back to your team immediately. Learn more about Callum's session: https://lnkd.in/gCmg-8sU Grab your tickets to DDD Melbourne now before they sell out! https://lnkd.in/gC4yyEbF

11

Is technical debt grinding your engineering organisation to a halt? If features are slow, bugs are piling up, and your team is struggling, you're not alone. Emily Bache explains that technical debt, a term coined by Ward Cunningham, describes when your internal code design falls out of sync with current business needs. It's crucial to understand that technical debt is NOT the same as bad code. Bad code is inherently complex and hard to change, while technical debt is about a mismatch between the current design and the design you now realise you need. Many common approaches to technical debt, such as large-scale rewrites or relying on "cleaning crews" of expert refactorers, often fall short or introduce new problems. Furthermore, while generative AI tools are helpful for many coding tasks, current Large Language Models (LLMs) are not reliable for refactoring, correctly performing in only 37% of cases in studies and frequently making designs worse or introducing bugs. This is because they operate on text, not the program's underlying structural parse tree. So, what is the most effective strategy? The video highlights that the best approach is to enable the developers who ordinarily work with the code to refactor it as part of their job. This involves: • Utilising proven, deterministic tools often built into IDEs (like 'extract function' or 'rename'). • Providing training and coaching for developers on essential refactoring skills. • Implementing a policy of "preparatory refactoring" or "tidy first," where developers improve code design before building new features. • For heavily indebted codebases, expert coaches can guide and empower ordinary developers, rather than just taking over. Stop letting technical debt impede your progress and frustrate your team! Learn the strategy that keeps your code flexible, your releases continuous, and your team productive. 🔗 Watch the full video now and transform how you manage technical debt: https://lnkd.in/e7VZVNti #TechnicalDebt #SoftwareEngineering #Refactoring #ModernSoftware #ContinuousDelivery #DevOps #Programming #SoftwareDevelopment #TechLeadership

6

When I ran a 70 person engineering org for a $1B company our software development cycle was simple and predictable: product, engineering, and design would spend a few weeks in meetings speccing out requirements, understanding customer pain points, and developing a solution. Then the team would go heads down for a few months building, iterating, and polishing. That ratio made sense - a few weeks of planning for a few months of building meant planning was maybe 10-15% of the total project time so it was worth investing in getting it right. That math is completely broken now. You can still have all the meetings and the discovery sessions and the alignment calls but what happens when your team can turn around and deliver working code in a day or two using the current set of AI tools? The meetings and coordination become the bottleneck. What used to take up a small percentage of the total time is now the majority of the time. It's like spending an hour debating where to go for lunch when you’re just getting Wendy’s and devouring it in 15 minutes. I'm seeing different ways teams are adapting to this. One model is that the product manager just does a lot more - they prototype, work on the UX, share it around, and only hand it off to engineering once it's pretty far along and ready for productionalization. Another model is to embrace much shorter release cycles - build with imperfect information and iterate rapidly based on real feedback, knowing that your ability to fix problems is now fast enough that you don't need to plan for perfection. What has to change is the entire process of developing software. The how of writing code got dramatically faster but we're all still figuring out the what we're building and why we're building it. The companies that figure this out are going to have a huge advantage over those still running the old playbook. This is also why AI is best done in house right now. Consultants just reintroduce a worse versiin of the coordination problem. They need to build relationships, understand the business, etc - all the things that are now the bottleneck. To really take advantage of AI you need people who live and breathe your business.

83

32 Comments

⏱️ Aligning QA with Project Timelines Delivering quality on time requires more than testing at the end. At 29FORWARD Australia, we help teams embed QA from project kickoff, not just pre-launch, ensuring fewer surprises and stronger results. Here’s what makes the difference: ✔️ QA aligned with delivery milestones ✔️ Early test design (shift-left approach) ✔️ Continuous collaboration between PMs and QA ✔️ Test environments ready when needed Say goodbye to last-minute defects and misaligned expectations. Say hello to on-time, high-quality delivery. 💬 Let’s talk about integrating QA into every phase of your project lifecycle. #QAStrategy #ShiftLeft #ProjectDelivery #SoftwareQuality #TestPlanning #29FORWARDAustralia #DevOps #QualityEngineering #AgileDelivery

4