Building a Semantic Model Referee With Google ADK and the OpenAI Agent SDK

Author(s): Kyle knudson

Originally published on Towards AI.

A practical comparison of two emerging agent frameworks for enterprise data engineering

A practical comparison of two emerging agent frameworks for enterprise data engineering

Over the last year, the momentum behind agentic AI has shifted from conceptual demos to real engineering workflows, with data teams moving beyond dashboards and RAG bots to build goal-driven, tool-using agents that enhance code evaluation, data quality, and governance. This article introduces the Semantic Model Referee, an agent created to evaluate semantic model YAML files, score their quality, and identify issues through a project that was built twice using different frameworks: Google’s Agent Development Kit and OpenAI’s Agent SDK. It shares insights into the architectural differences, the advantages of each framework, and the implications for data engineering teams needing automated quality checks and governance enforcement.

Read the full blog for free on Medium.

Join thousands of data leaders on the AI newsletter. Join over 80,000 subscribers and keep up to date with the latest developments in AI. From research to projects and ideas. If you are building an AI startup, an AI-related product, or a service, we invite you to consider becoming a sponsor.

Published via Towards AI