AI/ML Engineer focused on production machine learning systems, healthcare data platforms, MLOps, and practical AI engineering workflows.
I work across Python, Databricks, PySpark, Delta Lake, MLflow, model validation, data quality, and AI/ML observability. My engineering focus is building workflows that are reliable, traceable, and easy for teams to operate.
- Production ML and data pipelines
- MLOps and model lifecycle systems
- Databricks and Spark optimization
- Healthcare data and scoring workflows
- MLflow-based model versioning and tracking
- AI observability and evaluation-oriented workflows
- Codex-assisted context engineering and developer productivity
Languages: Python, SQL
Data & ML: PySpark, Databricks, Delta Lake, MLflow, Snowflake, Airflow
Engineering: Pytest, Poetry, Nox, GitHub Actions, developer tooling
AI/ML: scikit-learn, TensorFlow, Keras, BERT, NLP, model validation
Current Interests: AI observability, LLM evaluation, agentic workflows, MLOps/LLMOps
Building and supporting internal platform capabilities for model workflows, experiment tracking, reproducibility, data quality, lineage, and production operations.
Designing and improving large-scale data and model workflows involving claims, clinical data, risk scoring, readmissions, and future-cost use cases.
Improving preprocessing and inference workflows through Spark profiling, caching, broadcast joins, query refactoring, and cluster tuning.
Exploring model monitoring, evaluation, traceability, and structured context workflows for better AI-assisted development.
- Arize and AI/ML observability
- OpenTelemetry and tracing concepts
- LLM evaluation workflows
- Agentic AI development patterns
- Better ways to structure engineering knowledge for AI-assisted work
- Portfolio: https://vaibhavs825.github.io
- LinkedIn: https://www.linkedin.com/in/vaibhavs825
- GitHub: https://github.com/vaibhavs825
- Email: vaibhavs825@gmail.com