Thomas Talks AI

Production Claude Code evals should not begin with abstract benchmarks. Start with the agent runs that scared you, reduce them into replayable cases, and use them to tune permissions, prompts, tools, and review gates.

Claude Code permission modes can look safer than they are. The real production risk lives in tool scope: paths, network access, secrets, deploy files, and what reviewers actually approve.

If a Claude Code agent changes production code, the useful artifact is not the chat transcript. It is a flight recorder: intent, boundaries, commands, diffs, tests, approvals, and rollback notes.

A Claude Code demo is easy to love. You describe a feature, the agent edits files, runs commands, fixes its own mistakes, and suddenly the repository has moved. The first time you see it work, it feels like software engineering has skipped a generation. But the demo is not the hard part. The hard part is making that same capability safe enough, repeatable enough, and observable enough that you would trust it inside a real engineering workflow. That is where the actual product begins. ...

It was a Tuesday, the kind where the London drizzle seemed to seep into your bones, even if you were miles away in Basingstoke (my home town). I was tasked with a rather daunting project, or not really as you will find out: designing and deploying a real-time Sentiment Analysis product for a major telecommunications company, and, oh, they needed it yesterday. Or so it felt. Now, usually, this would mean weeks of planning, coding, debugging, and endless cups of tea. But this time, something was different. I decided to try something I’d been experimenting with – something I now call “Vibe Coding.” ...

Imagine you’re building a bridge. Not just any bridge, but one that needs to handle the daily rush hour of a major city. You wouldn’t use flimsy materials, would you? You’d choose steel, concrete—materials known for their strength and reliability. Now, think of your AI application as that bridge. What language will you choose to ensure it can handle the traffic of real-world use? Java or Python, or perhaps something more exotic or would you let AI decide for you? ...

Hey everyone! So, we’re on a massive hiring spree for AI Engineers at the AI labs (HCLTech), globally. And as someone leading the AI Engineering team, I’m trying to really nail down what makes a great AI Engineer. But honestly? It feels like we’re all speaking different languages when it comes to one key concept: AI Agents. Let me give you some background. I’ve been around the block. Years ago, I was deep into Java, building low-latency, high-performance cloud apps. Then, I dove into machine learning at Lloyds, training models for fraud and analytics. Now? We’re building AI and GenAI products and MVPs, pushing the boundaries. ...

Introduction Why isn’t AI revolutionizing business as expected? Despite the widespread use of tools like ChatGPT and customer service bots, the actual productivity gains seem minimal. Could it be that we’re not using AI the right way? AI’s true potential lies beyond basic tasks like writing emails or answering simple queries. To truly harness it, we need to reframe our approach, shifting from AI as a mere tool to AI as a co-worker that can transform operations. ...