AI, Software Engineer
As coding agents become more capable, the valuable skill shifts toward direction: defining the task, setting boundaries, reviewing output, and owning the decision.
Engineering notes from the sharp end
Practical writing about software, systems, delivery, and things that behave strangely under pressure.
wade womersley – york based software engineer I write the way I work: direct, useful, and more interested in what holds up in production than what sounds clever on a slide.
More posts
More posts from the archive
AI, Software Engineer
Claude Sonnet 4.6 is a reminder that model choice is becoming less about prestige and more about matching cost, latency, context, and task difficulty.
AI, Programming
Gemini’s recent tooling updates are another sign that agent development is becoming an orchestration problem, not just a prompt problem.
Programming, Software Engineer
Node.js moving toward one major release per year should make production planning simpler for teams that already care about LTS stability.
AI, Software Engineer
AI tools are now part of the software supply chain. That means they need the same security scrutiny as any other tool with access to systems and secrets.
AI, Software Engineer
Agent platforms are starting to compete on the plumbing: harnesses, deployment, monitoring, auth, and the boring parts between demo and production.
AI, Software Engineer
The more agents use real tools, the more they need boring infrastructure: isolation, versioning, profiles, credentials, and repeatable setup.
AI, Software Engineer
Developer documentation is becoming an interface for AI agents as well as humans. That means clean markdown, metadata, and tool access matter more.
AI, Programming
Next.js is starting to treat AI agents as real users of the framework. That is more important than it first sounds.
Programming, Software Engineer
TypeScript 7.0 Beta is interesting because the feature is performance. Faster typechecking and editor feedback can change how a large project feels.