Google Cloud BlogChristina Lin2 min readintermediate
The DevFest Community Workshop Experience: Building Real Agents Together
Summary
Google’s DevFest Community Workshop introduced a “Workbench” format that emphasizes architectural mental models over copy‑paste code, guiding engineers to build long‑running, self‑evolving multi‑agent systems with the Agent Development Kit and Gemini Enterprise platforms. Attendees learned state‑separation, workflow pausing, and self‑patching pipelines, and the series will continue in five more c…
- Workbench shifts focus from copy‑paste code to mental models like separating state from compute for long‑running agents.
- Labs used Google’s Agent Development Kit, Gemini Memory Bank, and RAG Engine to build self‑evolving multi‑agent systems with pause/resume workflow control.
- Participants integrated BigQuery data into autonomous pipelines, adding deterministic bidding logic and self‑patching harnesses for spend‑anomaly detection.
- Fast‑paced quizzes reinforced concepts and encouraged networking across GDG chapters.
Engineers building LLM‑driven agents need practical patterns for state management, cost control, and safe self‑patching; this workshop showcases concrete approaches.
5/10
