01
LLM Tooling Framework
A modular toolkit for building, testing, and observing useful model-driven applications.
AI Engineering / Coding / Personal Lab
I build with curiosity and rigor.
Turning ideas into working systems.
Documenting the process.
A modular toolkit for building, testing, and observing useful model-driven applications.
Practical workflows that combine AI assistance with human judgment instead of replacing it.
Small, reliable pipelines for data collection, processing, analysis, and operational cleanup.
Local-first knowledge management and semantic search experiments with simple interfaces.
LLM applications
RAG & Agents
Evaluation
Python / TypeScript
APIs & Services
Testing & CI
Data pipelines
Databases
Cloud & DevOps
Prototyping
Experiment design
Technical writing
What actually matters once a prototype needs to become something another person can trust.
How I move from idea to working system without letting the stack become heavier than the problem.
Why retrieval quality, workflow ergonomics, and observability usually matter more than louder claims.
Small systems survive longer when they are clear enough to adapt without accidental complexity.