Projects
Six projects across the Python data and AI stack — from ETL pipelines to multi-agent systems.
Data Pipeline Automation Script
PlannedA CLI tool that ingests raw data from files or APIs, validates and transforms it with pandas, then loads it into a SQLite or PostgreSQL database. Full schema validation, logging, and pytest coverage.
FastAPI Data Service
PlannedA production-ready REST API built on top of the data pipeline database. Supports filtering, pagination, and aggregation endpoints with full Pydantic validation and Swagger docs. Dockerized for deployment.
ML Model Training & Serving Pipeline
PlannedAn end-to-end machine learning pipeline: data preprocessing, model training with experiment tracking via MLflow, and a FastAPI prediction endpoint. Reproducible and deployable.
Agentic Data Analyst
PlannedAn AI agent that accepts natural language questions about a dataset and autonomously writes SQL queries, runs pandas operations, generates charts, and explains the results. Built with LangGraph and Streamlit.
Automated ML Pipeline with Orchestration
PlannedA scheduled ML pipeline that detects new data, retrains the model, compares it against production, and auto-promotes if performance improves. Orchestrated with Prefect.
Multi-Agent Research & Report System
PlannedA four-agent system where specialized AI agents collaborate to produce structured research reports. A Researcher, Analyst, Writer, and Reviewer work in sequence via LangGraph with a Streamlit interface.