Manal Rizwan

I’m a Data Scientist who builds AI-powered analytics and machine learning systems end-to-end—from data ingestion and modeling to production-ready reporting. I’ve architected and shipped an AI Data Agent that enables secure natural-language querying across Postgres and BigQuery using hosted MCP services and a Claude plugin.

In my recent roles, I’ve delivered decision-driving analytics for Delivery Operations and Customer Experience, built NL→SQL and dbt lineage sync, and implemented secure exports to signed GCS links so query data doesn’t enter model context. I also develop ELT pipelines with Google Datastream → BigQuery and dbt transformations that power company-wide dashboards and reporting.

Previously, I worked on loan monitoring and repayment risk, improving early default detection and forecasting repayment risk using statistical, ML, and deep learning models (LSTM, CNN, TCN, Transformers). I also build customer segmentation and dynamic NLP-based expense scoring, and I translate results into stakeholder-ready insights using Power BI, SSMS, and Python.