I build the data layer AI runs on: Python pipelines, LLM tools and MCP servers that turn the messy web into AI-ready data.
- 30+ data systems in production, 15+ open-source data tools
- Top Rated on Upwork, 100% Job Success
- Background in offensive security, so I design for failure and abuse from the start
- Currently learning: fine-tuning and evaluating open models on data I collect myself
Core stack: Python, FastAPI, PostgreSQL, Redis, Celery, Docker, Playwright, curl_cffi, LLM APIs (OpenAI, Anthropic), MCP
| Project | What it does |
|---|---|
| tubescrape | YouTube search, channels, playlists and transcripts without an API key. On PyPI, and the engine behind a YouTube data tool on the Apify Store |
| linkedinscrape | Turns LinkedIn profiles into structured data for profile analysis |
| shopscout | Product and collection data from any Shopify store |
| reddipy | Reddit posts, comments and user activity as structured data |




