How Novy Built Zelma for Emily Oster Using OpenAI and Postgres

Novy, a consultancy firm, collaborated with economist Emily Oster to develop Zelma, an AI-driven educational app designed to make U.S. state assessment data accessible. The project entailed overcoming data accuracy and query speed challenges, optimization using AI, and a robust stack including GCP, Kubernetes, and Postgres.

• Emily Oster is an economist at Brown University.
• Zelma includes data for Grades 3-8 from all US states.
• Novy developed a 24-hour prototype using Python and SQL.
• Query speed optimizations reduced response time to under 5 sec.
• Future plans include possibly switching to ClickHouse database.