|
Mathematician · AI/ML Engineer · Managing Partner at Collective AI I build AI systems that move from research to production, with a current focus on agentic software engineering, generative AI, voice AI, local LLMs, and applied machine learning. I also enjoy working at the intersection of mathematics, deep learning, developer tooling, and chess. |
|
- 🤖 Agentic engineering — workflows, skills, governance, and tooling for coding agents.
- 🧠 Applied AI & ML — production systems, forecasting, deep learning, and generative AI.
- 🗣️ Voice AI — real-time assistants where speech, reasoning, and actions stay synchronized.
- 🖥️ Local LLMs — practical ways to run capable models on constrained hardware.
- 📚 Math & Deep Learning — educational material connecting mathematical foundations with modern architectures.
| Project | What it is | Stack |
|---|---|---|
| Orquesta Lite | Local, durable workflow runtime for coordinating coding agents with verified flows, quality gates, and resumable execution. | Go |
| Botica | Agentic engineering toolkit for bootstrapping repositories and sharing reusable coding-agent skills and standards. | Python |
| Math Deep Learning | A five-volume collection covering mathematical foundations, advanced architectures, reinforcement learning, NLP, and graph deep learning. | TeX |
| Qwen Local | Run Qwen locally with llama.cpp + Docker, including configurations aimed at GPUs with limited VRAM. |
Shell / Docker |
| Engineering Skills | Reusable agent skills that encode production engineering standards for refactoring and PR review. | Python |
| Voice Chess Assistant | A voice chess coach that speaks, analyzes with Stockfish, and moves pieces in sync with its narration. | Python / React |
AI Agents · LLMs · Machine Learning · Deep Learning · Voice AI · Forecasting · Developer Tools · Python · Go · Mathematics
