Course Objective & Scope
While inline autocomplete speeds up individual functions, real engineering time goes into exploring codebases, coordinating changes across dozens of files, running test suites and fixing build errors. This course teaches how to direct terminal and IDE coding agents with rigorous engineering discipline.
Participants practise read-only exploration and plan mode before execution, closed-loop terminal/test/linter verification with Claude Code, Cursor Agent, OpenAI Codex CLI and Windsurf, connecting Jira/GitHub/databases/browsers via MCP (Model Context Protocol), and running parallel agents with git worktree on a real repository.
Who Should Attend?
- Experienced software engineers (mid-level, senior, staff)
- Tech leads and software architects
- Platform and DevEx engineers
- Prerequisite: hands-on fluency with the command line (CLI), Git and automated testing tools
Key Takeaways
- Turn complex engineering tasks into verifiable implementation plans before any file is edited
- Run agents in closed loops that search, edit, compile and execute tests autonomously
- Connect agents to docs, issue trackers, database schemas and Playwright via MCP servers
- Prevent context rot in long sessions by compacting state and splitting tasks cleanly
- Run multiple independent tasks in parallel without conflicts using Git worktrees and subagents
Curriculum
Coding Agent Architecture and Tool Ecosystem
- Comparing Claude Code, Cursor Agent, Codex CLI and open-source coding agents
- Autonomy levels: when to use chat, plan mode or full autonomous execution
Exploration, Plan Mode and Task Decomposition
- Having the agent perform codebase discovery and impact analysis before editing
- Producing a reviewable implementation plan with explicit verification checkpoints
Closed-Loop Execution: Terminal, Test and Linter Feedback
- Using compiler, typechecker and unit-test output as self-correction signals
- Managing the context window with
/compact,/clearand state files
Model Context Protocol (MCP) and External Tool Integration
- Connecting GitHub/GitLab, Jira/Linear, PostgreSQL and live library docs via MCP
- Verifying UI changes visually and functionally with Playwright/browser MCP servers
Parallel Engineering with Subagents and Git Worktrees
- Delegating research, implementation and verification to isolated subagents
- Running concurrent tasks safely in separate
git worktreedirectories
Other Courses in This Category
-
End-to-end product-to-engineering workflow training that turns business requirements, PRDs, user stories and Figma designs into technical specs, task slices, acceptance criteria and verified code.
- Live Course
- 4 modules
-
Advanced training on building execution harnesses around coding agents—fast test/typecheck/lint feedback, custom tool wrappers, AST/LSP verifiers and internal eval benchmarks—so agents succeed reliably in enterprise repos.
- Live Course
- 5 modules
-
Advanced training on designing closed-loop autonomous coding cycles (generate → compile → test → observe logs/traces → repair) with convergence criteria, budget caps, anti-reward-hacking rules and nested loop architectures.
- Live Course
- 5 modules
-
Combining static analysis metrics with LLMs to understand unfamiliar codebases, detect technical debt and code smells, execute safe refactoring backed by characterization tests, and optimise performance.
- Live Course
- 5 modules
-
Developer training on generating unit, integration, contract and property-based tests with AI, using mutation testing to verify that generated test suites actually catch defects rather than mirroring implementation bugs.
- Live Course
- 5 modules
-
Solving complex software bugs through hypothesis-driven debugging with logs, stack traces, OpenTelemetry traces, git bisect and minimal reproducible examples (MREs) instead of blind trial-and-error prompting.
- Live Course
- 5 modules
Get in touch
