I build AI agents so I have someone to talk to while debugging.
Based in 🏠 Working from home | 150+ ⭐ | Building the future of AI tooling
I bridge the gap between deep AI/RAG architectures and production-ready full-stack delivery. Because being "Full-Stack" wasn't exhausting enough, I decided to add Machine Learning into the mix.
- 🧠 LLMs & Agents: Building assistants, tool-use workflows, and evals. (Translation: Teaching rocks to think, then adding guardrails so they don't say anything unhinged).
- 📚 RAG & Search: Vector DBs, chunking, and hybrid retrieval. (Translation: Forcing AI to actually read the documentation before it answers).
- ⚙️ Infra & Backend: Temporal, AWS Lambda, FastAPI, Django. (Translation: Orchestrating cloud functions and connecting pipes before they orchestrate a rebellion).
- 🎨 Frontend & Mobile: Next.js, Flutter, 3D rendering. (Translation: Making sure the AI's output actually looks good to human eyes).
I am language-agnostic and pick the right tool for the job. Here is what I'm usually spinning up:
Frontend, Mobile & Cloud Infrastructure
Plus: Pinecone, Weaviate, FAISS, LangChain, LlamaIndex, Temporal, and OpenTelemetry.
Whether it's web, mobile, or pure backend infrastructure, I like to experiment:
- 🤖 Agentic Workflows: Built complex AI shopping assistants utilizing LangGraph, Groq, and Streamlit.
- ☁️ Cloud Orchestration: Mastered durable execution by orchestrating AWS Lambda functions with Temporal in TypeScript.
- 📄 Document Intelligence: Developed multiple PDF QA / RAG proof-of-concepts.
- 📱 Cross-Platform: Shipped a sleek, neumorphic music streaming app with local storage support in Flutter/Dart.
- 🧊 Browser Graphics: Created Silkcards-3D, analyzing Adobe Illustrator files to render them in browser-based 3D.
If you want to talk about RAG architectures, complain about Vector DB pricing, or build something cool, my DMs are open.





