Real-World Multi-Agent Development | Ship Faster with AI Agents
summary
Demonstration builds a URL shortener using ASP.NET Core API, Razor Pages front end, SQL Server, Azure deployment and GitHub Actions while orchestrating work with Squads multi-agent orchestration and the GitHub Copilot CLI. The system casts a five‑to‑seven member agent team, scaffolds a repo (about 128 files created), auto-generates 21 issues (15 appear quickly), and produces an architecture doc and CI workflows; agents pick models (Claude Opus 4.6 for design, Sonic 4.5 for reviews) and require permission tuning to avoid bad commits and markdown escaping artifacts.
AI-written summary · about
chapters (14)
- 0:00Multi-agent development intro
- 1:06Squad orchestration
- 2:31Setup Squad and Copilot CLI
- 4:17Initialize Squad team
- 6:25Cast team with prompt
- 9:56Approve team creation
- 13:34Project structure
- 16:26Agent roles and charters
- 19:41Solution architecture
- 22:34Team directives and workflows
- 25:04GitHub issues generation
- 27:13Fix formatting issues
- 28:38Review issues and steps
- 29:23References and closing
related talks
52mCut the Hype and Build Your Own AI Agents
01h 08mFrom Model Integration to Agents: Building Intelligent .NET Applications
56mBuilding AI Agents and Multi-Agent Systems with Azure AI
18mCollaborative AI Engineering: One Dev, Two Dozen Agents, Zero Alignment
01h 04mAgentic AI: Build a Multi-Agent Application with CrewAI
58mAdvanced Autonomous AI Agents: Memory, Collaboration and Evolution in Action
05h 17mBuilding AI Applications
26mAI powered automation & multi-agent orchestration