HAMZA MESKINI

I’m an AI Engineer specializing in AI Systems and Generative AI, focused on building production-ready LLM-powered automation pipelines. I prioritize decoupling cognitive AI processing from deterministic logic to eliminate hallucinations, maximize architectural scalability, and deploy high-ROI industrial software.

In my recent internship at Stellantis (Global Test & Validation Division), I designed and deployed an AI platform on Databricks adopted by cross-border validation teams, cutting test design and configuration time by 90%. I architected a deterministic hybrid resolution engine (Python, SQLite, Pydantic) to map and validate inputs against 12,935 DCI signals, and I built end-to-end generative pipelines (FastAPI, LangChain, Claude) with ChromaDB RAG plus MLflow-tracked LLM evaluation. I also developed supporting tooling across the stack—Jinja2-based test module generation, FastAPI backend APIs, and React/TypeScript frontends—so stakeholders could align outputs with ControlDesk, ecu.test, and CANoe constraints.