About
I build intelligence from real data.
I work at the seam of data engineering and machine learning — the production pipelines that move real-world data, and the AI systems built on top of them. The most ambitious of those is sarpOS — a personal life-AI I'm building to learn, remember, and reflect. (Vasuki, the guide you met on the home page, is the friendly face on top of it.)
What I do
I design data infrastructure that doesn't break overnight, models a business can trust, and code other engineers can read — then I push past the pipeline into the ML stack: retrieval, evaluation, agents. The goal is always the same: turn messy, real-world data into systems that scale.
Tools I work with
Data & pipelines
- Apache Spark
- PySpark
- dbt
- Airflow
- Kafka
- Delta Lake
Cloud & platforms
- Azure
- AWS
- Databricks
- Snowflake
AI / ML
- PyTorch
- LangChain
- HuggingFace
- RAG
- LLMs
- Vector DBs
- MLflow
Languages
- Python
- SQL
- TypeScript
Vasuki — the long game
Vasuki is my attempt at a genuinely personal AI — not a wrapper around a chatbot, but a system that learns, remembers, and acts within boundaries I set. It's early, it's in development, and it's the throughline of everything else here: every pipeline and model is practice for building a mind I can trust.