Field notes on how AI tools like Cursor and Claude are reshaping software engineering—honest observations from a profession in transition.
It’s been three months since I last wrote a line of code. I still deliver features. I still solve problems, tackle architecture challenges, review PRs. I’m not a PM. I’m not a tech lead with a team of agents. I still consider myself a developer. But I am not doing the same job I was doing a year ago. The gradient, not the cliff There was no single moment where I thought…
I know that feedback about using AI as a dev is so mid-2025, but I feel there are still some things to say. I just built a query builder interface in ~2 days - 5k lines of code, complex UI, not a small chunk of work. It’s a playground where users can build queries across multiple data sources with advanced filters. Think simplified Metabase. Good candidate for AI assistance, as I knew where…
Most leverage from AI doesn’t come from writing AI code—it comes from writing code that AI can use. When your system exposes clear contracts and examples, assistants can extend it safely. Building a copilot and an MCP server at my job made this obvious: after we added a contract API and short, focused docs, “write-a-prompt” replaced “write-a-spec” for many changes.…
Seven hours on a plane, zero internet, and a curiosity about local AI coding. That’s how I found myself downloading 13GB of AI models in an airport lounge, preparing for what would become an accidental experiment in offline development. This wasn’t a research project or a formal comparison of existing solutions. I was heading back from vacation with a long flight ahead and wanted to…
Developers on my team use Cursor’s Tab feature. Designers use vibe-coded prototypes to gather feedback from stakeholders. I rely on Claude Code to plan and build larger features. Same AI technology, completely different approaches. The conversation around AI coding tools assumes a binary: you either use AI or you don’t. But that misses how practitioners actually work. In my journey…
When I first started using Claude Code, I had a naive approach to working with it. I would describe the task directly in the prompt, press Enter, and cross my fingers. If the agent made mistakes, I would tell it how to fix them. For small tasks, this can be good enough, but as the task grows in complexity, this approach reveals several significant drawbacks. When Simple Doesn’t Scale The…
For the first time in my career, I have absolutely no idea what the software engineer job will look like in five or ten years. I’ve always loved coding. All of it. I love making apps from scratch as much as understanding big, complex codebases. I take pleasure in taking a complex problem and finding a simple implementation, or taking a complex piece of code and making it simpler. This love…