CodeScene

Enabling AI coding assistants to request Code Health insights directly from a codebase. Let's see how devs and AI-coding assistants can collaborate to improve code quality with MCP server. Use cases with MCP server: 🚀 Safeguard new code: Prevent AI from introducing technical debt by flagging code health issues like excess complexity, low cohesion and so on. 🚀 Targeted Refactoring: AI tools can refactor code, but they lack direction on what to fix and how to measure if it helped. The MCP Server largely solves the problem by giving the AI precise insight into design problems. 🚀 Understand existing code: LLMs have their flaws, but they are generally great at summarizing information. We can use this to create reports and diagnostics from all the detailed code health reviews. #aicoding #technicaldebt #codequality #codehealth

🤖 Can AI really understand the code it writes? At Øredev yesterday, Adam Tornhill and Markus Borg took on this question and revealed new insights into how AI could reason about its own code and evolve it safely. It’s not enough for AI to generate code, it also needs to understand what good code looks like. 💡 Defining “good” code in the era of AI is just the beginning. And on that note, join us for our upcoming webinar with Codility “Beyond correctness: what makes AI-generated code ready for production?” Sign up for the webinar: https://lnkd.in/dSX2wYNW #aicoding #ai #codequality #codehealth #softwareengineering

🚀 Join CodeScene’s upcoming webinar to learn practical tools and techniques for crafting maintainable, high-quality code - especially in the era of AI-assisted development. 🗓 Wednesday, November 26 | 🕓 4–5 pm CET | 💻 Online P.S. Get a sneak peek of CodeScene’s MCP Server, and see how AI agents can start improving code quality using real Code Health insights from your codebase. 🎙️ Speakers: Kalle Norrestam, Tech Lead, CodeScene Selma Čopra, Software Engineer, CodeScene 👉 Sign up for the webinar https://lnkd.in/gCPKKXjN #technicaldebt #codequality #aicoding #codehealth

Writing code with AI is easy. Trusting it in production? That’s a different story. On Nov 19, Codility and CodeScene are teaming up to share brand-new research on what really makes AI-generated code ready for the real world. We’ll go beyond surface-level accuracy and dig into: 🔧 1. Maintainability 🔐 2. Engineering trust 📈 3. Long-term code quality This one’s for engineering leaders, hands-on devs using AI and anyone thinking about how to future-proof their software workflows. 📌 Save the date - Wednesday 19th. 🔗 Link to register is in the comments.

🚀 WEBINAR INVITE: How do you really know if AI-generated code is ready for production? Passing correctness tests only tells one part of the story. Sign up for the webinar: https://lnkd.in/dSX2wYNW Join us for a webinar where experts from Codility and CodeScene will talk about groundbreaking research that revolutionizes how we evaluate AI-generated code. What You’ll Hear: ✅ How AI-generated solutions that are technically correct can still fail due to inefficiency, poor maintainability, or complexity that creates technical debt. ✅ How Codility’s COMPASS benchmark evaluates three critical aspects of AI-generated code: correctness, efficiency, and quality. ✅ What the research means for teams using AI coding assistants, including which models show the most consistent performance and reliability. Sign up for the webinar: https://lnkd.in/dSX2wYNW #aicoding #codequality #technicaldebt

Ready for some new AI research? We all know AI can write code, but can it write good code? Adam Tornhill and Markus Borg are doing a session at Øredev this Thursday, where they'll also reveal some exciting new research. 🚀 As Adam puts it: "We all know AI can write code, but can it write good code? And, more importantly, will the AI be able to reason about its own code and evolve it safely? We promise, we'll define what 'good' means." Don't miss the session. #aicoding #codequality #technicaldebt #ai

How one week changed everything. How do you turn slowing development into real progress? When Peter Caron joined HelloFresh, development had hit a wall, too many bugs, tangled dependencies, and teams struggling to stay aligned. 💡 Within his first week, Peter introduced CodeScene, identifying the exact areas holding them back and answering the big questions from leadership: Why are we moving slower? Why are we seeing more bugs? Why are our teams struggling to collaborate? Watch how data-driven insights helped bring clarity and control back to the codebase and set the foundation for faster, healthier development. #technicaldebt #codequality #codehealth

How long would it take you to grasp 40k lines of legacy C++, all crammed into a single class? My AI message over these past few years has been that I see plenty of promise, *but* we're focusing AI on the wrong problem. (IMHO, of course). Typing was never the bottleneck in programming. Human comprehension is. (And no: today's AI is nowhere near being a "replacement" for programmers). Now, with MCP servers, we're finally seeing what real AI-assisted coding looks like. Within seconds, I had Copilot summarizing a ridiculously complex class. Better, I also get an actionable refactoring guide that factored in risk. Is AI + MCP the ultimate UX for working with legacy code?

How can we help AI coding assistants understand code health and improve code quality? Tune in and find out, we’re building something special. 🚀