Projects
Selected architectures, pipelines, and backend systems I have built.
ETL Pipeline
An end-to-end orchestrated data pipeline that processes and integrates 5.8 million flight records with weather data from an API source. Deployed on GCS utilizing a Medallion data architecture for the downstream flight delay risk prediction models.
Python • Pandera • GCS • Airflow • Docker
RAG Engine
A Retrieval-Augmented Generation system designed for querying technical documentation. The architecture features a Qdrant vector store for semantic search and LLM ingestion. It utilizes a FastAPI backend and Pydantic data validation, integrated with an SQLite database schema for efficient local memory management.
Python • Qdrant • FastAPI • Pydantic • SQLite