Agentic Software Engineering using GitHub Copilot

This four-day workshop takes you from GitHub Copilot fundamentals to full agentic software engineering. Across nine modules you will build expertise across the complete Copilot ecosystem, covering agent mode basics, the Copilot harness of instructions, MCP servers, skills, plugins and hooks, agentic coding with local agents, cloud agents and subagent orchestration, agent sessions and the Agents Window, the Copilot CLI and SDK, the Copilot desktop app, agentic DevOps, governance, cost and observability, and planning and spec-driven delivery. Every concept is grounded in hands-on practice with real tools and real scenarios.

The way software gets built is changing fast. AI agents are no longer just productivity tricks: they are becoming core members of the engineering team. This course is built for developers who want to lead that shift while their peers are still catching up. You learn to delegate the repetitive, the predictable and the mechanical to agents, so your own attention goes toward decisions that need judgment, context and creativity.

You start with GitHub Copilot fundamentals and Agent Mode, then assemble the harness that agents run on, implement agentic coding with local agents, cloud agents and subagent orchestration, and manage agent sessions in the Agents Window, including remote sessions over SSH and dev tunnels, before automating with the Copilot CLI and SDK. Advanced modules cover the Copilot desktop app for session-based issue delivery, agentic DevOps for infrastructure and pipelines, governance, cost and observability, and the Plan, Specify, Deliver module, where you harvest a brownfield system, plan the change with agents and ship a feature through GitHub Spec Kit. By the end you have a working mental model for agentic systems, a set of reusable patterns you can bring back to your team, and the confidence to apply them immediately.