Hi, I'm Usama. Tech Lead on the GenAI team at Educative, where I've spent 4+ years across three roles slowly becoming the person everyone asks about LLMs.
My work lives at the intersection of prompt engineering, multi-agent systems, and RAG pipelines. I worked on Fenzo AI — a multi-agent content generation system — from the ground up. I also make interactive ML/AI demos that actually explain things instead of just looking cool (though they look cool too).
I have strong opinions about AI-generated code and I will die on that hill.
Outside work: Philosophy, Game Theory, and following the GenAI space obsessively.
P.S. Currently deep in AI Evaluation, Context Engineering, and Token Efficiency. Trying to get better at all three simultaneously. Not recommended.
- LLMs & Prompting — Prompt architecture, multi-agent pipelines, RAG, evals
- Frameworks — LangChain, LangGraph, FastAPI, CrewAI
- Vector Stores — Pinecone, Chroma
- ML & Training — PyTorch, fine-tuning, quantization (learning)
- Cloud & Infra — AWS / GCP, Docker
- Viz & Demos — Three.js, D3.js
- Languages — Python, JavaScript, Go
- Tech Lead, GenAI — Educative (Current)
- Senior Software Engineer, GenAI — Educative
- Software Engineer — Educative
