At RadarSoft, I owned a Snowflake warehouse consolidating finance, operations, and time-tracking data for reporting. I built Python and SQL ELT pipelines from three source systems and added ingestion validation that kept accuracy above 99%.
I also cut query latency 40% by examining query profiles, restructuring joins, and consolidating repeated logic into a materialized step. Before that, as a Data Analyst at iConsult Collaborative, Syracuse University, I prepared environmental and geographic data from public sources for comparable analysis across 50+ countries.
At the School of Information Studies, Syracuse University, I built an Airflow-orchestrated data pipeline feeding a retrieval system, with Postgres marts, pgvector embeddings, and automated evaluation using hit@k and MRR. I containerized the stack and deployed its query service behind FastAPI, with GitHub Actions running dbt builds and tests on every push.
