1-Day Instructor-Led Training
Microsoft Official Content
Hands On Labs
Microsoft Certification Exam Voucher Included
After-Course Instructor-Coaching Included
Developing in Agentic AI Systems (GH-600)
Course 8764 Advantage Plan Course
- Duration: 1 day
- Language: English
- Level: Intermediate
This course teaches learners how to develop, deploy, and manage agentic AI systems within GitHub-based software development workflows. Participants learn how AI agents can be integrated into the software development lifecycle, including agent architecture design, tool and environment configuration, memory and state management, execution, evaluation, and governance.
Learners explore how to operate agent workflows inside the SDLC, supervise autonomous behavior using GitHub controls, evaluate and tune agent outputs, configure custom agents, and coordinate multi-agent execution safely.
Agentic AI Development in GitHub Course Delivery Methods
In-Person
Online
Upskill your whole team by bringing Private Team Training to your facility.
Agentic AI Development in GitHub Course Information
Course Benefits
- Builds practical skills for developing and managing AI agents in GitHub workflows
- Helps development teams integrate agents safely into the SDLC
- Covers agent architecture, tooling, MCP servers, execution environments, memory, state, and evaluation
- Supports responsible use of AI agents through governance, guardrails, and human oversight
- Helps teams supervise autonomous agent behavior while maintaining traceability and control
- Prepares learners for the GitHub Certified: Agentic AI Developer certification
Prerequisites
Learners should have experience with the software development lifecycle, GitHub workflows and controls, code quality practices, security practices, and code review processes. They should also have experience with coding agents such as GitHub Copilot, MCP servers, custom instructions, custom agents, tools, and Copilot setup steps.
Exam Information
Agentic AI Development in GitHub Course Outline
Foundations of Agentic AI in GitHub
In this section, learners explore how AI coding agents are changing software development by planning, acting, and improving within GitHub workflows.
- Define agentic AI in the SDLC
- Explain the agent lifecycle: plan, act, evaluate
- Describe GitHub as the system of record and control plane
- Identify responsibilities, risks, anti-patterns, and traceability needs
- Apply the contributor model to agent-generated work
Agent Architecture and SDLC Integration
In this section, learners design agentic systems that use GitHub workflows to build software safely, reliably, and with appropriate controls.
- Map agent responsibilities to the SDLC
- Define inputs, outputs, and success criteria
- Separate planning, reasoning, and execution
- Apply pull request governance using templates, checks, CODEOWNERS, rules, and environment gates
- Build reliable workflows with outputs, contexts, triggers, and cross-job handoffs
- Control and operate agents using observability, tools, MCP, secrets, hooks, and reliability practices
Tooling, MCP, and Agent Execution Environments
In this section, learners configure the tools, execution environments, and boundaries agents need to perform tasks safely within GitHub workflows.
- Explain how agents interact with GitHub APIs and workflows
- Use Model Context Protocol servers, registries, and allow lists
- Define execution context and boundaries
- Apply agent execution limits and protections
Multi-Agent Systems and Orchestration
In this section, learners design reliable multi-agent systems in GitHub using observable workflows, coordinated artifacts, and safe recovery mechanisms.
- Design multi-agent workflows for coordinated development tasks
- Define agent roles, responsibilities, and handoff points
- Coordinate artifacts across agents and workflows
- Monitor multi-agent activity for reliability and traceability
- Apply recovery mechanisms when agent workflows fail or require intervention
Memory, State, and Evaluation
In this section, learners manage agent memory and state, persist progress across environments, and evaluate agent behavior using clear success signals.
- Manage agent memory and state across tasks
- Persist progress across development environments
- Define evaluation criteria for agent outputs
- Use scans, artifacts, and signals to assess quality
- Tune agent behavior based on evaluation results and error analysis
Governance, Guardrails, and Operations
In this section, learners design secure and compliant agent governance using GitHub-native controls, human approvals, least-privilege access, and operational safeguards.
- Implement governance and guardrails for agentic workflows
- Configure human-in-the-loop approvals
- Apply least-privilege access controls
- Improve accountability through logging, review, and traceability
- Use operational safeguards to support reliability, recovery, and safe production use
Need Help Finding The Right Training Solution?
Our training advisors are here for you.
Agentic AI Development in GitHub Course FAQs
This course is designed for developers, AI engineers, DevOps engineers, platform engineers, solution architects, GitHub administrators, and technical leads who want to develop, supervise, and govern AI agents in GitHub-based software development workflows.
No, this as an intermediate course. Learners should already understand GitHub workflows, repositories, branches, pull requests, CI/CD concepts, and basic AI fundamentals.
The course covers agent architecture, SDLC integration, GitHub workflows, MCP servers, execution environments, agent memory and state, evaluation, error analysis, multi-agent orchestration, governance, guardrails, and operational controls.
Yes. Learners should have experience with coding agents such as GitHub Copilot, and the course focuses on developing and managing agentic AI systems within GitHub-based workflows.
Yes. This course aligns with the GitHub Certified: Agentic AI Developer certification.