Fernando Wittmann

I’m a Senior Machine Learning Engineer and applied Data Scientist with 8+ years of experience building production-grade ML and NLP systems on messy, real-world data. My work sits at the intersection of data pipelines, model evaluation, and LLM-powered applications — with a strong bias toward reliability, measurability, and deployment over research prototypes.

In recent roles, I’ve designed end-to-end systems for extracting structured signals from unstructured documents (contracts, PDFs, policy-like text), built evaluation frameworks for NLP and LLM workflows, and deployed embedding-based retrieval and RAG pipelines used in production. I’ve worked extensively with Python, Pandas, SQL, vector search, and cloud infrastructure (AWS), and I’m comfortable defining precision/recall-driven metrics for entity extraction, classification, and agent behavior.

I care deeply about trustworthy AI: deterministic evaluations, structured outputs, bias awareness, and systems that hold up under real operational constraints. I’m comfortable explaining and leveraging transformer internals — including attention — to improve prompt design, context construction, and debugging in applied LLM systems.

I thrive in small, senior teams, enjoy collaborating closely with engineering and product, and prefer roles where data science directly shapes the platform rather than living in isolation. I’m excited by Stitch’s focus on production AI, evaluation rigor, and solving hard problems in a compliance-sensitive domain like insurance.