For engineering managers, tech leads and staff engineers who must roll out autonomous coding agents, keep quality and security intact, and measure whether it is working.
The curriculum is being prepared and reviewed. Enrollment is not open yet.
About this course
Coding assistants have become coding agents: tools such as Claude Code, OpenAI Codex, GitHub Copilot's coding agent (and GitHub Agent HQ, which runs several of them side by side) and Cursor's cloud agents can take an issue, change many files, run tests and open a pull request. That changes the job of an engineering leader. This course is about the decisions that follow: which work to delegate to agents, how to make a codebase agent-ready with AGENTS.md or CLAUDE.md instructions, fast tests and clear conventions, and how to change code review when a large share of diffs were not written by a person.
You will also work through the risks that matter at team level: secrets and production credentials in agent sandboxes, prompt injection through issues and dependencies, permissions for MCP servers and tools, license and provenance questions, and the quiet erosion of junior engineers' skills. The course closes with measurement, using the DORA AI Capabilities Model and delivery metrics to separate real throughput gains from rising instability, and with a phased rollout plan. Tool capabilities and plan availability are described as of October 2026 and change frequently.
What you will be able to do
Compare the major AI coding agents on autonomy, execution environment, integration and governance controls, and select one for a given team
Prepare a repository for agent work with instruction files, reliable tests, CI gates and scoped permissions
Decide which tasks to delegate to agents and write issue specifications agents can complete reliably
Adapt code review, ownership and merge policies for agent-authored pull requests
Identify and mitigate security risks from coding agents, including secret exposure, prompt injection and over-privileged MCP tools
Measure the impact of agent adoption with delivery and stability metrics rather than self-reported speed
Plan a phased rollout covering training, career development for junior engineers and cost governance
Syllabus
6 modules · 20 lessons · about 12 hours
1
Module 1: From Assistants to Agents
What changed when coding tools began running commands and opening pull requests, and what that means for the team.3 lessons
Autocomplete, Chat, Agent: Three Levels of Delegation
The 2026 Agent Field: Claude Code, Codex, Copilot Coding Agent, Cursor and Agent HQ
Local, Cloud and CI Agents: Where the Code Actually Runs
2
Module 2: Making the Codebase Agent-Ready
The repository investments that decide whether agents produce mergeable work.3 lessons
Instruction Files: AGENTS.md, CLAUDE.md and Copilot Custom Instructions
Tests, Linters and Fast CI as the Agent's Feedback Loop
Writing Issues an Agent Can Finish: Scope, Acceptance Criteria and Context
3
Module 3: Review, Ownership and Workflow
Changing how the team reviews, merges and owns code when agents write much of it.4 lessons
Reviewing Agent Pull Requests: What to Read, What to Run, What to Reject
Ownership and Accountability: The Human Who Merges Owns the Change
Parallel Agents, Worktrees and Avoiding Merge Chaos
AI-Assisted Code Review Bots: Useful Second Reader or Noise
4
Module 4: Security and Compliance for Coding Agents
Keeping agents from becoming the easiest way into your systems.4 lessons
Sandboxes, Network Egress and Keeping Secrets Out of Agent Context
Prompt Injection via Issues, Docs and Dependencies
MCP Servers and Tool Permissions: Least Privilege for Agents
Licenses, Provenance and Audit Logs for Generated Code
5
Module 5: Measuring What Agents Change
Evidence-based measurement of throughput, quality and cost.3 lessons
DORA Metrics and the DORA AI Capabilities Model
Why Perceived Speed Misleads: Designing a Fair Before-and-After Comparison
Token Spend, Seat Costs and Premium Requests: Cost Governance for Agent Usage
6
Module 6: People, Skills and the Rollout Plan
Growing engineers, not just output, and rolling agents out in stages.3 lessons
Junior Engineers and Deskilling: Deliberate Practice in an Agent World
Hiring and Interviewing When Candidates Use Agents
A 90-Day Phased Rollout: Pilot, Guardrails, Scale
Exam and certificate
The exam unlocks once you have completed every lesson. It draws 20 questions from a larger bank, spread across the modules, with the answer options shuffled for each attempt. You have 45 minutes and need 80% to pass. After a failed attempt you can try again after 24 hours, up to 3 attempts in total. Feedback is given per module, never as the answer key.
Passing earns a certificate of completion and assessment from the Wirewalk Institute of AI with a unique ID that anyone can check at institute.wirewalk.com/verify. It is not an accredited qualification or a vendor certification. How certification works.
The Wirewalk Institute of AI is independent. This course teaches publicly documented products and is not affiliated with or endorsed by GitHub, Anthropic, OpenAI, Cursor, Google, or Microsoft. Product and company names are trademarks of their respective owners.
Practical AI for finance and accounting: reviewing models in Excel with Copilot and Claude, reconciliation and close support, invoice extraction, FP&A commentary, and the SOX and audit-evidence discipline that keeps AI-assisted numbers reliable.
Using AI in legal work without ethical or evidentiary missteps: ABA Formal Opinion 512 and the Model Rules, grounded research in Westlaw and Lexis, contract review, eDiscovery, court AI-disclosure orders and building a practice-group AI policy.
AI across the employee lifecycle with the legal guardrails built in: drafting and screening in recruiting, NYC Local Law 144, Illinois HB 3773 and the EU AI Act's employment rules, handbook assistants, performance writing, and employee privacy.