I'm an enthusiast of AI and software development, currently learning by building. I've been interested in this field for a long time — reading about it, experimenting, and slowly turning what I learn into small projects.

I'm not claiming to be an expert. I'm in the middle of the journey and enjoying it. Most of what I do is about understanding how AI systems actually work, end to end.

• VoiceLaunch TTS — a local-first desktop text-to-speech app for assistive communication: type text, hear it spoken, and route the audio to playback or a virtual mic. (Electron, React, Python/FastAPI)
• LumenAI SDK — a small SDK I made to explore cost tracking and observability for GenAI usage. (Python, OpenTelemetry)
• CapiNews Data — automated daily AI news/report data with a simple static frontend.
• AI Engineering Notes — my study notes on AI engineering, RAG, LLMs and MCP.

As part of learning, I've started contributing to a few open-source projects (like Qdrant) — small, well-scoped fixes with tests. Reading real codebases and getting feedback from maintainers has been one of the best ways for me to improve.

Python, TypeScript, FastAPI, React, and anything around RAG, LLMs and voice AI. Mostly I'm just curious, and trying to get a little better each week.