Christopher Brennan

Designed and launched end-to-end AI systems using Databricks, Python, and LLMs to support VA benefits modernization, including veteran claims and benefits processing features. Built RAG pipelines with LangChain, FastAPI, and ChromaDB for context-aware Q&A grounded in VA policy and eligibility guidance. Implemented semantic search using OpenAI embeddings with Elasticsearch and Pinecone to improve retrieval precision. Orchestrated Databricks ML pipelines using PySpark, Delta Lake, MLflow, and Databricks AutoML for auditable experimentation and deployment in FedRAMP-aligned environments. Deployed LLM-powered chatbots and multi-agent workflows using LangChain/LangGraph to assist case workers. Enforced governance and security using Databricks Unity Catalog, including field-level encryption, role-based access controls, and audit logging; delivered inference APIs with FastAPI integrated with React dashboards.

Developed predictive analytics for insurance claim risk and eligibility scoring using TensorFlow 1.x, Keras, and scikit-learn. Ensured HIPAA compliance and FHIR-standard interoperability by applying encryption and anonymization during data preprocessing. Built healthcare NLP pipelines using spaCy, NLTK, and embeddings for entity extraction and automated note summarization. Deployed batch ML workflows on AWS EC2 and on-prem systems using Airflow and S3 for artifacts. Created ETL and preprocessing with Apache Spark (MLlib) and Pandas; delivered REST inference services with Flask and uWSGI. Applied models such as Random Forests, SVMs, and gradient-boosted trees for fraud detection and payment forecasting, and integrated OCR using Tesseract/OpenCV for digitization and classification. Used hyperparameter tuning (GridSearchCV, early stopping) and managed AWS resources (EC2, RDS, Lambda, CloudWatch, S3) for scalable workloads.

Built real-time writing assistance and grammar correction microservices using Python, Kafka, and AWS. Designed event-driven streaming infrastructure with Apache Kafka to support instant text analysis and writing suggestions. Migrated legacy applications to Python services and added Redis-based API caching. Used AWS services (EC2, ECS, Fargate, Lambda, RDS, CloudWatch) to scale systems. Developed web dashboards and reporting tools using React/Redux integrated with real-time data streams. Implemented REST APIs with Flask and Django REST Framework, including Swagger/OpenAPI documentation and JWT authentication. Improved early NLP/ML grammar correction using TensorFlow, scikit-learn, and NLTK through training classifiers on large text corpora.