Measuring what a local LLM actually costs on a GPU-less machine.
- Python
- Typer
- Pydantic
- Docker
- GitHub Actions
Portfolio · Systems, local AI, tooling
Computer science and electronics student. Multi-profile AI agents, model routing, cybersecurity and benchmarks on modest hardware. Everything here is public, tested and documented — not slides.
Work
What I build for myself is the best evidence of what I can build for you. Public code, tests and documentation included.
Measuring what a local LLM actually costs on a GPU-less machine.
Running a multi-agent system at home, without the cloud, and keeping control of the data.
This showcase: static, bilingual, no CMS and no database.
About
That constraint — doing a lot with few resources — is exactly the one most small organisations face. Which is why I favour sober, measurable solutions you can take over without me.
Contact
A question about a project, an idea for collaboration, feedback to share? Write to me, I reply within 24h.