ML Engineer at CVS Health. Five years taking ML from raw data to production — large-scale data engineering and lakehouse architecture, forecasting and causal inference, multi-agent LLM systems, and the MLOps to ship and run them reliably.
Built to ship at production scale — not prototypes that stall in notebooks.
| Project | What it shows |
|---|---|
| safe-model-deployment | Champion–challenger deployment: traffic splitting, sequential testing, automated rollback |
| model-observability | Production model monitoring: PSI/KL drift detection, latency SLOs, automated alerting |
| ml-platform-demo | End-to-end ML platform: training pipelines, model serving, monitoring, CI/CD on Kubernetes |
| gpu-ml-learning | GPU programming with Triton and CUDA through hands-on ML examples |
Interactive documentation sites I build and maintain while studying — each one turns a dense technical book or topic into a structured, navigable reference.
ML systems & MLOps — mlops-daily-dose · ml-feature-store · distributed-machine-learning · ai-systems-performance · gpu-tutorial
Data engineering — ddia · fundamentals-de · de-design-patterns · practical-lakehouse-architecture · streaming-db · database-internal · kafka-tutorial · apache-flink
Systems & infrastructure — scalable-systems · operating-systems · how-linux-works · k8s-docs-tutorial · fundamentals-of-devops · opentelemetry · high-performance-python
Python SQL PyTorch Spark LangChain/LangGraph Vertex AI Airflow Dagster MLflow BigQuery Snowflake Docker Kubernetes Terraform
Ship it. Monitor it. Scale it.