Public proof of AI systems work: evaluation-first retrieval, grounded generation, and a smaller set of technical experiments worth showing.
Carta Healthcare · Data Scientist
Owned clinical ETL and NLP pipelines at Carta Healthcare, standardizing heterogeneous healthcare inputs into HL7/FHIR-aligned workflows and integrating LLM-assisted extraction into registry operations.
TripleTen · Data Science and AI/ML Instructor
Built a rigorous mentoring and review practice at TripleTen, guiding learners through end-to-end ML systems with emphasis on evaluation, failure analysis, and production-minded technical judgment.
Standalone flagship build · AI Systems Builder
Built an evaluation-first RAG system for dense technical documents, using a labeled benchmark and three retrieval baselines to separate retrieval quality from grounded answer quality before optimizing generation.
Applied machine learning workflows · ML Engineer and Data Scientist
Built structured evaluation workflows for model comparison, validation rigor, and error analysis so production ML decisions were driven by evidence instead of surface-level metrics.
Applied AI and knowledge workflows · AI Systems Builder
Designed PDF-based retrieval systems with careful chunking, grounding, evaluation, and failure analysis so answers stayed useful beyond the demo stage.