Genai Applied
Sign in Join the waitlist

AI for your profession · Software Engineering

AI for software engineers, built around the code you already ship.

Coding agents now write, test and review real code. This course teaches you to direct them across a whole repository, and to build the AI features your product team is asking for.

Or get your personal AI playbook now

What changes

Your work, re-mapped.

How AI is changing this work today, task by task. Your assessment maps it onto your own week.

Tasks AI can already do for you.

  • Boilerplate, scaffolding and CRUD endpoints
  • First drafts of unit and integration tests
  • Docstrings, READMEs and migration notes
  • Routine dependency upgrades

The curriculum

Ten modules.
One path through them.

Your assessment orders these for your level and goals and skips what you already know. Open any module to see what it covers.

01How coding agents actually work5 topics

What happens between your prompt and a merged pull request, so you can predict where agents fail.

  1. Tokens, context windows and why agents forget
  2. Tool calls: read, edit, run, search
  3. Terminal agents vs IDE agents vs cloud agents
  4. Where agents hallucinate APIs and how to spot it
  5. Choosing a model for the task
02Context engineering for your repo5 topics

Give the agent the knowledge a senior teammate would have on day one.

  1. Project memory files and conventions docs
  2. Scoping tasks so they fit in context
  3. Feeding specs, schemas and examples
  4. Keeping secrets and private code out of prompts
  5. Measuring when context helps or hurts
03Agentic coding workflows5 topics

Plan, implement, test and iterate with an agent in the loop, not just autocomplete.

  1. Plan-then-execute prompting
  2. Small diffs and reviewable steps
  3. Running several agents in parallel with worktrees
  4. Delegating tickets to cloud agents
  5. When to take the keyboard back
04AI-assisted testing and QA5 topics

Use AI to raise coverage and catch regressions without drowning in flaky tests.

  1. Generating tests from specs and bugs
  2. Property-based and edge-case discovery
  3. Making AI-written tests trustworthy
  4. AI for test data and fixtures
  5. Visual and end-to-end testing with agents
05Code review with AI5 topics

Make AI the first reviewer, and keep humans on the decisions that matter.

  1. Automated PR summaries and risk flags
  2. Security and dependency checks
  3. Review checklists an agent can follow
  4. Avoiding rubber-stamp reviews
  5. Team norms for AI-written code
06Debugging and incident work5 topics

Find root causes faster with an agent that can read logs, traces and code together.

  1. Reproducing bugs with an agent
  2. Reading stack traces and logs at scale
  3. Hypothesis-driven debugging prompts
  4. AI in on-call and postmortems
  5. Knowing when the agent is guessing
07Extending agents with MCP and tools5 topics

Connect agents to your databases, tickets and internal APIs safely.

  1. What the Model Context Protocol is
  2. Using existing MCP servers
  3. Writing a small MCP server
  4. Permissions and least privilege
  5. Prompt-injection risks from tool output
08Building AI features into your product5 topics

Ship LLM-powered features: chat, search, extraction and summarisation.

  1. Calling LLM APIs well: streaming, structured output
  2. Retrieval-augmented generation basics
  3. Latency and cost budgets
  4. Fallbacks when the model fails
  5. Privacy and data handling
09Evaluating AI output5 topics

Prove a prompt or feature works before users find out it doesn't.

  1. Building a small eval set
  2. Assertions, rubrics and LLM-as-judge
  3. Regression testing prompts
  4. Tracing and observability
  5. Shipping changes behind evals
10Your AI-native engineering practice5 topics

Turn all of this into a repeatable way of working for you and your team.

  1. Measuring real productivity gains
  2. Keeping your own skills sharp
  3. Security, licensing and IP questions
  4. Rolling AI out across a team
  5. A 30-day plan for your codebase

Outcomes

By the end, you can.

  • 01Hand an agent a real ticket and review its pull request with confidence
  • 02Set up project context so agents follow your team's conventions
  • 03Connect agents to your own tools through MCP
  • 04Ship and evaluate an LLM-powered feature in your product

Tools and techniques you'll use

  • Claude Code
  • Cursor
  • OpenAI Codex
  • GitHub Copilot
  • Model Context Protocol (MCP)
  • Git worktrees
  • LLM APIs (OpenAI, Anthropic, Gemini)
  • Evals and tracing

Who it's forDevelopers, tech leads and QA engineers who want to work with agents, not around them.

The course is opening soon

Be first in.

Join the waitlist for the Software Engineering course. One email when it opens, nothing else.

Questions

Good to know.

Will AI coding agents replace software engineers?

They replace a lot of typing, not engineering. Teams still need people who decide what to build, design systems, review what agents produce and own production. Engineers who can direct agents well get more done; this course is about becoming one of them.

Which tools does the course use?

The workflows apply to terminal agents like Claude Code, IDE agents like Cursor and cloud agents like Codex. We teach the patterns behind them so you aren't tied to one product.

When does the course open?

The full course is being built now and opens to the waitlist first. Join with your email and we'll tell you as soon as it's ready.

How is the course personalised?

Before the course, a five-minute assessment, by text or by voice with Iris, our AI guide, learns your role, daily work, skills and goals. Your personal playbook then orders these modules for your level and goals and skips what you already know.

What can I do while I wait?

Take the assessment now. You get your personal AI playbook straight away, with the tasks in your work AI can take over, the ones it speeds up, and your learning path.