itsAlex31
AI engineer, 4 years total: three years of full-stack web development, and since late 2025 working full-time on LLM agent systems.
Most of my work is on the parts of an agent nobody sees. A large share of it is deciding what tools an agent should have in the first place — researching the gaps, prototyping, and building the ones that turn out to matter — since an agent is mostly only as capable as the tools you hand it. The rest is the infrastructure that keeps a long-running agent coherent: summarizing context so it doesn't lose the thread over a long session, and condensing logs into something the agent can actually reason over instead of drowning in. Python, TypeScript and Go, across both commercial and open-source models.
I use coding agents heavily in my own workflow — writing specs, giving them context, reviewing what comes back, and owning what ships.
Looking for remote AI/Software Engineer positions. Small teams and high ownership suit me. Open to relocation.