Home › Certifications › GitHub Agentic AI Developer (GH-600)
Microsoft · DevOps

GitHub Agentic AI Developer (GH-600)

Exam GH-600 (Developing in Agentic AI Systems) earns the GitHub Certified: Agentic AI Developer certification. Engineers who deploy, operate, supervise and govern AI agents inside production software delivery, using GitHub as the system of record and control plane: configuring custom agents, evaluating their output and coordinating multi-agent runs safely.

Exam code
GH-600
Certification
GitHub Certified: Agentic AI Developer
Level
GitHub certification · Advanced
Duration
100 minutes
Passing score
700 / 1000
Exam cost
USD 99 (US price, varies by country)
Renewal
Time-limited; renew per GitHub policy
Difficulty
Challenging

Quick Verdict

A GitHub credential that shows practical skill with the platform. It is worth it for engineers and teams that live in GitHub every day. Plan your study around 'Implement tool use and environment interaction', which carries 20–25% of the exam.

Skills Measured

Weightage shows the approximate share of questions from each area.

Prepare agent architecture and SDLC processes

15–20%
  • Integrate agents into the SDLC
  • Boundaries between planning, reasoning and action
  • Observability and control

Implement tool use and environment interaction

20–25%
  • Agent tools
  • MCP servers
  • Development environments
  • Safe execution and error handling

Manage memory, state, and execution

10–15%
  • Memory strategies
  • Persist state and manage context drift
  • Continuity across tools

Perform evaluation, error analysis, and tuning

15–20%
  • Success criteria
  • Root-cause analysis
  • Tune behaviour

Orchestrate multi-agent coordination

15–20%
  • Multi-agent workflows
  • Observability
  • Failure handling
  • Agent lifecycle

Implement guardrails and accountability

10–15%
  • Autonomy levels
  • Guardrails and human-in-the-loop

Overview

Exam GH-600: Developing in Agentic AI Systems is the exam you pass to earn the GitHub Certified: Agentic AI Developer certification from GitHub, delivered through Microsoft Learn and Pearson VUE.

Engineers who deploy, operate, supervise and govern AI agents inside production software delivery, using GitHub as the system of record and control plane: configuring custom agents, evaluating their output and coordinating multi-agent runs safely.

The exam is split into 6 skill areas. The heaviest is Implement tool use and environment interaction at 20–25% of the questions, so that is where most of your preparation time should go.

Skills listed on this page follow Microsoft's official study guide for this exam.

Exam format

You get 100 minutes for the exam. Microsoft scores exams from 1 to 1,000 and you need 700 or more to pass. Expect multiple-choice and multiple-response questions, plus formats such as drag-and-drop, case studies and hot-area items. You can sit the exam at a Pearson VUE test centre or online with a remote proctor. If the exam is not offered in your language, you can ask for 30 extra minutes.

How to earn it and keep it

Production experience with GitHub workflows and AI coding agents.

GitHub certifications are time-limited. Check the renewal rules on the certification page in Microsoft Learn before the expiry date.

Who This Suits

Engineers who deploy, operate, supervise and govern AI agents inside production software delivery, using GitHub as the system of record and control plane: configuring custom agents, evaluating their output and coordinating multi-agent runs safely.

Prerequisites

Production experience with GitHub workflows and AI coding agents.

Career Paths

AI-assisted Software EngineerAgent Platform EngineerSenior DeveloperDevOps Engineer

Frequently Asked Questions

Which certification does exam GH-600 lead to?

Passing exam GH-600 (Developing in Agentic AI Systems) earns the GitHub Certified: Agentic AI Developer certification.

How long is the GH-600 exam and what is the passing score?

You get 100 minutes to complete the exam. The passing score is 700 / 1000.

Is GitHub Agentic AI Developer (GH-600) worth it?

A GitHub credential that shows practical skill with the platform. It is worth it for engineers and teams that live in GitHub every day. Plan your study around 'Implement tool use and environment interaction', which carries 20–25% of the exam.

What topics does the GitHub Agentic AI Developer (GH-600) exam cover?

Prepare agent architecture and SDLC processes (15–20%); Implement tool use and environment interaction (20–25%); Manage memory, state, and execution (10–15%); Perform evaluation, error analysis, and tuning (15–20%); Orchestrate multi-agent coordination (15–20%); Implement guardrails and accountability (10–15%).

How much does the GH-600 exam cost?

USD 99 (US price, varies by country). Microsoft sets exam prices by country or region, so the price shown at checkout is the one that applies to you.

Does the certification expire?

Time-limited; renew per GitHub policy.

What jobs can GitHub Agentic AI Developer (GH-600) lead to?

Common roles include AI-assisted Software Engineer, Agent Platform Engineer, Senior Developer, DevOps Engineer.

Verify before you book: exam fees, formats and prerequisites change from time to time. Confirm the current details on Microsoft's official study guide before you register.