USC, Viterbi School of Engineering
AMAR
A multi-agent RAG pipeline that turns plain-language questions into analytical reports.
Sep 2024 – Jun 2025
From a plain-language question to a finished analytical report in five minutes
A multi-agent RAG pipeline on Databricks and Airflow that parses a question, plans the analysis, and writes the report end to end, cutting turnaround from roughly an hour of analyst work to about five minutes across three domains. Retraining runs through the MLflow lifecycle rather than manual redeploys, and I load-tested it at 50M events/day on replayed clickstream before trusting the number.
- ~1 hr → 5 min
- report turnaround
- 50M
- events/day in load test
- 3
- business domains served
- Databricks
- Airflow
- MLflow
- RAG
- Multi-Agent Systems
- Python
The problem
Non-technical stakeholders needed custom analytical reports but had to route every request through an analyst, costing roughly an hour per report and gating insight behind a person's queue.
What I built
- 01Built a multi-agent RAG pipeline that parses a plain-language question, plans the analysis, and produces a custom report end to end.
- 02Orchestrated the pipeline on Databricks with Airflow DAGs so runs were scheduled, observable, and reproducible.
- 03Wired automated retraining and parameter tuning through the Databricks MLflow model lifecycle rather than manual redeploys.
- 04Stress-tested at 50M events/day using replayed public clickstream data before trusting the throughput numbers.
Outcome
Report turnaround dropped from roughly an hour of manual analyst work to about five minutes, across financial-reporting, healthcare-diagnostics, and customer-support scenarios.