Hassan Zahid

I’m a production-focused Machine Learning Engineer who thrives on forward-deployed AI—turning models into reliable services, fast APIs, and measurable business outcomes. I’ve built and supported ML systems at scale, including “900K+ daily requests” served with FastAPI and BentoML on Vertex AI.

I specialize in RAG and retrieval systems that work in real organizations. I productionized internal RAG on GCP for a “40,000+ employee organization,” including chunking, embedding generation, vector index population, metadata-filtered retrieval, and retrieval tuning.

I also build and operate production ML with rigorous performance signals. I supported “Saved $100K+” in daily fraud risk exposure by deploying and retraining LightGBM fraud detection models on Vertex AI batch inference, tracking metrics like PR-AUC and recall at fixed false-positive rates.

In parallel, I deliver LLM agent workflows with guardrails and testing. As a “Senior AI Engineer” contract, I translated stakeholder requirements into deployed agent behavior (prompt logic, tool usage, fallback handling), created “25+ QA scenarios,” and improved reliability for voice-agent edge cases. I’m equally comfortable with production release pipelines using Docker, Terraform, GitLab CI/CD, and GCP infrastructure.