Guy Royse

New Cypress blog post showing how to mock the locked down "location.href = ..." assignments in your E2E tests: "Mock The Location Href Property" at https://lnkd.in/eRjSYNfn

11

Grow engineering velocity without growing the security backlog. Build security into every commit so code ships safely, no matter whether it’s written by humans or AI.

3

We’ve been iterating quickly on our .NET testing agent, and some early patterns are becoming hard to ignore. Compared to what we shipped in 18.3, we’re seeing roughly a 4× improvement in end‑to‑end task completion on the same benchmark suite (generate, build, run, learn, iterate). The more interesting story, though, is efficiency. We’re not optimizing for “number of tests generated”. We care about how much useful signal we get from each run, and how much work actually turns into coverage instead of retries. Newer versions consistently deliver: - deeper coverage per run, especially across branches and edge cases - more coverage gained per unit of effort, with significantly lower token waste - fewer runs that stall early due to setup or infrastructure issues Across the same scenarios used by other agents in this space, what stands out is efficiency: on resolved tasks, we typically see around 1.5–2× stronger coverage lift per run, depending on the repository. Scale is another clear differentiator. As scope increases (larger solutions, longer‑running scenarios), many general agents start to degrade or stall. Our agent is designed to handle large scopes and long runs without fundamentally changing its behavior, which makes a real difference on bigger, more complex codebases. We’re actively looking for feedback, especially on what didn’t work. Repos where the agent stalled, struggled with scale, or produced low‑value tests are the most valuable signals for us to investigate and improve. If you’ve tried AI‑assisted testing on production .NET code: what broke for you? #dotnet #testing #unittesting #developerproductivity #ai #qualityengineering

12

3 Comments

Very excited! Improving has acquired DragonSpears, a software development company specializing in cloud adoption, agile coaching and DevOps services. Headquartered in Chicago, this acquisition will significantly bolster Improving’s current presence in the Midwest region, as well as increase Improving's annualized revenue beyond $285 million. More about the acquisition here: https://hubs.ly/Q03nc7-v0 Improving has offices in 19 cities spread across US, Canada, Mexico, Argentina, Chile and India. https://lnkd.in/gwFRhY5M

10

1 Comment

🎤 Chris McFee (KeyBank), Adam Parker (Huntington National Bank), and Darshana Pai (Virtusa) are set to take the {unscripted} stage in Columbus to discuss the evolution of DevOps practices, along with challenges and expectations. They will share how to effectively scale DevOps in complex, regulated environments by streamlining cloud migrations, aligning teams around DORA metrics, and fostering cultural change. 🌎 Not in Columbus? See the complete list of cities and our virtual event. 👉 https://lnkd.in/dhAKHU3B

1

Grow engineering velocity without growing the security backlog. Build security into every commit so code ships safely, no matter whether it’s written by humans or AI.

3

🤔 How important is making things visible in your organization? when information like tests results are there available for everyone collaboration and innovation increase! If you want to move faster 🏃‍♂️‍➡️ , start by removing some "wastes" from your SDLC with Testkube which allows you to Catalog, Execute, Troubleshoot, & Report automated testing from one centralized platform 🤯

3

1 Comment

The recent discussion on PHP Architect's "Community Corner" with Larry Garfield about the PHP Framework Interop Group (FIG) truly resonated with me. While not always front-page news, the work of groups like FIG is foundational, enabling the seamless collaboration and component interchangeability we often take for granted in modern PHP development. This isn't just about cleaner code; it's about reducing friction, accelerating development cycles, and ensuring long-term maintainability for businesses relying on PHP applications. From a consultant's perspective, embracing these standards (like PSRs) is a non-negotiable best practice. It fosters an ecosystem where libraries and frameworks can coexist harmoniously, leading to more robust, scalable, and adaptable systems. In my Laravel projects, this commitment to interoperability extends beyond PHP itself – it informs how I design APIs for React frontends, structure reusable components in Flutter, or ensure data consistency with Python backends. It’s a core philosophy that drives robust, scalable, and future-proof solutions, minimizing technical debt and maximizing business value across the full stack. What "behind-the-scenes" standard or group has had the most significant, yet often overlooked, positive impact on your software development projects? #PHP #Laravel #SoftwareDevelopment #TechConsulting #Interoperability

1

Custom software isn’t just tailored, it’s built to last. Off the shelf tools may seem quick or budget friendly at first glance, but they come with hidden limitations: 1. rigid structures that stifle growth, 2. costly integrations as your needs evolve, 3. and an uphill battle against generic features. We’ve seen it again and again, custom solutions future proof your business by adapting seamlessly to scale, integrating precisely where you need them, and delivering long term savings. So here’s the real question: Are you asking for something that grows with you or settling for what fits in a box? #CustomSoftware #Scalability #BusinessGrowth

7

What if coding agents could reliably implement your software? Not just for a single session... but for an entire project 🤩 I kept hitting the same wall: the agent would start strong, then drift, then hallucinate requirements I never mentioned. By session three, we'd lost the orientation completely. The problem wasn't the AI. It was me. I was "vibe coding"—prompting on instinct without structured constraints. So I explored a different approach: Spec-Driven Development. The idea: make specifications—not code—the primary artifact. You define what and why before the agent writes a single line of code. The specs become a "memory bank" of your intent. The agent references the memory bank and implements your intent. I tested three open-source frameworks (Spec-Kit, OpenSpec, AgentOS) to explore how Spec-Driven Development works. My takeaway? The specific tool matters less than the approach. Full breakdown of my learnings and experiences on my blog: https://lnkd.in/ek93HPnn #AICoding #AgenticAI #SoftwareEngineering #ClaudeCode #SpecDrivenDevelopment

27

22 Comments

Seniors see beyond code - they see systems. 🔹 How a single feature impacts the customer journey 🔹 How one API design affects every future integration 🔹 How one design decision changes team velocity for months Writing code is the easy part. Anticipating impact is where senior engineers earn trust.

3

One small thing I like to implement is feature flags. 🏳️ It can be a smaller lift. 🙌 When I show a team who has no or little developer experience their eyes light up when they know they can turn it on/off like a validation rule or flow. 🔥 These nice to haves help non developer or low code orgs have faith and trust in introducing code for their more complex automations. ⚙️ #salesforce #salesforcedeveloper #salesforceconsultant

8

Watching a well-known developer tooling company publicly reckon with its business model this week made something very clear to me. Many organizations don’t realize how much revenue quietly depends on users landing on specific parts of their website. * A product Q&A section that influences a few percent of conversions. * A fitment or compatibility tool that reduces purchase hesitation. * A buying guide that nudges the decision. * A blog post that introduces the product before the user ever sees a PDP. None of these are usually mission-critical on their own. They’re often written off as “only” one or two percent. But they all share something important: they only work if the user reaches the page. And combined they likely represent a percentage of revenue you can't ignore. AI is increasingly answering those same questions before a customer ever sees your Q&A, your guide, your calculator, or your content. The demand for this information won't change, but the access and path will. Developers may be encountering this first, but they won’t be the last. Over the next 12–24 months, the organizations that fare better will be the ones that understand exactly where this dependency exists and what replaces it when the click goes away.

4

https://lnkd.in/gDG_yQga I mentioned to several ex-coworkers a year ago that really good QA folks are going to be great at Specification Driven Development (SDD) as they can be the layer between the daydreamers and the end product. But, if you are a good developer and a great daydreamer, you can daydream reality if you implement SDD methodology.

2

Someone asked me recently what I've learned after 23 years in QA. I didn't answer with a tool. I didn't answer with a methodology. I said: every quality problem I've ever walked into was either a People problem, a Process problem, or a Technology problem. And the fix is completely different depending on which one it is. A team that doesn't own quality won't be saved by better automation. A governance model that doesn't scale will collapse under its own weight. An automation suite with no architecture will cost more to maintain than it saves. Get People right — build the culture, hire for mindset, make quality a shared identity. Get Process right — build the governance layer that sticks, not the most complete one. Get Technology right — build automation that reduces the maintenance burden, not adds to it. That's the lens I've used to build TCoEs from nothing, run multi-million-dollar programs for Fortune 500 clients, and now explore what AI actually changes about quality engineering. Not the hype version. The practical one. People. Process. Technology. In that order. Always. #QualityEngineering #QALeadership #TCoE #AITesting #EnterpriseQA

116

11 Comments

Event Sourcing. Is it a powerful architectural pattern or just overengineering in most systems? If your domain doesn’t legally require audit,and business rules rarely change, would you still choose Event Sourcing? At what point does history become the source of truth instead of just supporting data? Where do you draw the line? #EventSourcing #DDD #SoftwareArchitecture #SystemDesign #Microservices

One of the best presentations I've seen in a while. We often read about or discuss Team Topologies, DDD, or Wardley Mapping as separate concepts, but this session masterfully joins them into a single, cohesive strategy. Using the practical example of an online education system to show how to migrate from a legacy "big ball of mud" to a new modular architecture, while simultaneously reorganizing teams for a faster flow of change. One to watch more than once. Great work by Susanne Kaiser! https://lnkd.in/efdX8PER #WardleyMapping #TeamTopologies #DDD

191

2 Comments

RAG Implementation: The Daily GrindWorking with RAG (Retrieval-Augmented Generation) can feel like it's "ragging" you sometimes. The retriever pulls in relevant docs from your vector store, but chunking and indexing issues often lead to irrelevant results. Then the generator tries to make sense of it all, sometimes still hallucinating despite the context. Common pain points: Noisy retrievals lowering precision Generator ignoring key context Tuning embeddings and reranking for better performance When it clicks though, RAG delivers grounded, accurate responses. Keep tweaking—your pipeline will stabilize.#RAG #AI #MachineLearning #LLM

1