Christian Henning's CV
Experience
Principal Data/ML/AI Engineer, Overproof, Inc -- Remote
June 2023 – present
-
Architected and implemented a scalable data pipeline to collect and process unstructured restaurant menus, enabling automated matching of alcohol beverage products.
-
Developed and deployed a suite of ML/AI applications to automate document processing workflows and enhance data quality assurance.
-
Implemented a high-performance vector search engine using semantic embeddings to resolve product entities, achieving 95% matching accuracy.
Principal Data/ML Engineer, Sensentia, Inc -- New York City, NY
May 2016 – June 2023
-
Architected and built the complete backend infrastructure and API for an automated health insurance benefit inference engine, utilizing NLP to process thousands of complex health insurance plan documents.
-
Engineered and developed a high-throughput document ingestion system for PDF health insurance plans, optimizing parsing logic to improve processing speed by 50%.
-
Spearheaded the transition to ML-driven models for benefit extraction, successfully reducing manual data entry and verification efforts by 40%.
Lead Data Modeler, TransRe -- New York City, NY
Aug 2011 – May 2016
-
Established the engineering foundation for the Catastrophe Center of Excellence (CCE) as the founding engineer, architecting a critical reporting system used for regulatory compliance and S&P ratings.
-
Engineered and optimized complex internal and external reporting pipelines, achieving a 60% increase in generation speed through query tuning and architectural improvements.
-
Designed and implemented robust data models for catastrophe analysis and disaster recovery, directly enabling improved risk mitigation strategies.
Software Engineer, AIR Worldwide -- Boston, MA
Oct 2007 – Aug 2011
-
Developed and actively maintained the core Property Analysis Engine and natural disaster models, optimizing algorithms to boost calculation efficiency by 25%.
-
Pioneered the research and integration of GPU processing for disaster modeling simulations, significantly accelerating computation times for complex scenarios.
-
Led the strategic migration of the legacy codebase to a 64-bit environment, unlocking substantial system performance gains and memory capacity.
Skills
Languages: Python, SQL, C#, C++
Data Engineering: Pandas, NumPy, Dask, PySpark, Kafka, Apache Iceberg
AI/ML Frameworks: scikit-learn, spaCy, PyTorch, OpenAI/Gemini API, pgvector, ChromaDB, LangChain
Databases: Postgres, SQLite, SQL Server
Infrastructure: AWS, GCP, Docker
Front End: Django, Streamlit
Developer Tools: Git, GitHub, GitHub Actions, Jira, Confluence
Projects
Explainable Naive Bayes
Jan 2026 – present
A simple, explainable Naive Bayes text classifier focused on transparency and first principles, written in plain Python.
-
Train and classify text based on labels.
-
Explain results in table format.
-
Porter Stemmer, Stop Words, Bigrams, Rare Words Exclusion, Cross Validation.
-
Accuracy, Confusion Matrix, Cross Entropy.
Money Stuff Newsletter
Dec 2025 – present
End-to-end RAG pipeline and analysis engine for financial newsletters using LLMs and semantic search.
-
Loading and cleaning of newsletter data into SQLite database.
-
Named Entity Recognition (NER) & Timeline Analysis.
-
Topic Modeling & Clustering.
-
Semantic Search (Vector Search).
-
Q&A Chatbot (RAG).
-
Sentiment Analysis.