AI training for engineering teams
Make your engineering team AI-native.
AI training for engineering teams. Put Codex, Claude Code and coding assistants to work on your own development process, from a clear brief to tested changes and useful AI features.
Live, custom training · Remote-first · On-site by arrangement
A working change you can review
- Plan · Scope and repository standards
- Build · A focused patch and relevant tests
- Review · Checks, assumptions and next steps
Codex
Claude Code
GitHubLearn to build the workflow behind the result.
Built for engineering
Start with a real task.
Leave with a better way to do it.
These are starting points for a custom programme. Choose the workflows that matter to your team; we shape the sessions around your tools, experience and goals.
Example workflowTurn a product brief into a working prototype.
01 · Start with your tools
CodexClaude CodeGitHubYour brief, codebase, sample data and team standards.
02 · Apply your process
Plan the change, build with a coding agent and run relevant tests before review.
Build a reusable workflow or skill03 · Review the result
Working code, checks and a clear list of what to improve next.Your team checks and approvesPlus 10 more ways to put AI to work.
Open an example to see the inputs, process and output.
02Set up a repository-aware coding agent
Start withYour repository, conventions and developer setup.
Learn toDocument build commands, boundaries and examples so the agent has useful context.
Walk away withA repeatable agent setup your team can maintain.
Claude CodeCodexGitHub03Build a custom MCP connection
Start withAn approved internal API and a narrow use case.
Learn toDefine tool schemas, scope permissions and test errors before connecting an agent.
Walk away withA testable MCP server with documented access boundaries.
Custom MCPsClaude CodeCodex04Develop a retrieval-backed assistant
Start withApproved documents and representative user questions.
Learn toImplement retrieval, expose sources and evaluate cases where the answer is absent.
Walk away withAn assistant prototype with a documented evaluation set.
LangGraphGitHubClaude Code05Build a multi-step agent workflow
Start withA bounded task, tool interfaces and success criteria.
Learn toModel the steps, add approval points and test failure paths.
Walk away withA workflow that can be inspected and evaluated.
LangGraph
Hermes AgentCustom MCPs06Generate meaningful test cases
Start withA feature specification, known bugs and current coverage.
Learn toIdentify edge cases, write tests and check that failures detect real problems.
Walk away withTests a reviewer can connect to expected behaviour.
Codex
CursorGitHub07Review a change with AI assistance
Start withA pull request, requirements and repository standards.
Learn toCheck correctness, examine failure paths and validate findings before reporting them.
Walk away withA prioritised review with reproducible issues.
GitHub CopilotClaude CodeGitHub08Modernise a small part of a codebase
Start withA bounded legacy component and its behaviour contract.
Learn toPlan an incremental change, preserve behaviour and test before widening the scope.
Walk away withA reviewable migration with rollback notes.
CursorCodexGitHub09Evaluate an AI feature
Start withRepresentative inputs, expected behaviour and unacceptable failures.
Learn toBuild an evaluation set, compare versions and inspect regressions.
Walk away withA decision-ready evaluation report with known limitations.
LangGraphCodex
Google Sheets10Turn engineering standards into skills
Start withYour review checklist and examples of accepted work.
Learn toPackage instructions and helper scripts, then test on a real task.
Walk away withVersioned skills the engineering team can improve together.
Claude CodeCodexHermes Agent11Prepare an incident investigation
Start withApproved logs, the timeline and system documentation.
Learn toGather evidence, distinguish hypotheses from facts and draft recovery options.
Walk away withAn investigation brief for the on-call engineer to verify.
GitHubClaude CodeCustom MCPsTools & connections
A broader toolkit.
Built around your stack.
Your programme can combine AI tools with the systems your team already uses. Pick the right starting point, then learn how to connect the pieces.
You do not need every tool here. We agree software access, available integrations and technical depth before the sessions. Logos identify example curriculum tools, not partnerships.
Connect your context
Approved documents, CRM records, project tools and internal data, using native connections or custom MCPs where appropriate.
Teach your process
Turn your playbooks, templates and checks into reusable prompts, skill files and workflow steps.
Keep people in control
Review the output, handle exceptions and decide what can be automated after the workflow has been tested.
Your programme
Learn it together.
Use it in your work.
For software engineers, technical leads, engineering managers and ai builders. Start with a focused workshop series or build capability through a longer custom programme.
What your team develops
- Coding-agent context, planning and code review
- Custom MCP servers, retrieval and agent workflows
- Evaluations, test design and repeatable engineering skills
- 01 · Scope
Choose the work
Map one priority workflow, the starting skills and what good output looks like.
- 02 · Practise
Build with an instructor
Work through live examples, get feedback and practise in your approved tools.
- 03 · Apply
Keep a reusable system
Document the workflow, its skills and review steps so your team can keep improving it.
Define success before you start
Measure the work,
not the tool count.
Agree a baseline and compare like-for-like tasks after training. For engineering, useful measures might include:
- Time to a reviewed, tested change
- Review rework and regressions
- Performance on a defined evaluation set
These are proposed measures, not promised improvements or customer results.
Learn with practitioners
The people behind
your programme.
Practical instruction, live feedback and a curriculum shaped around the work your team needs to do.
See the work behind the training
Inside a Modern Health AI workshop
Before you book
AI training for engineering.
Your questions, answered.
Your next step · 15-minute call
What could your engineering team
do differently?
Bring one workflow and the tools you use. We'll discuss a practical starting point, the right format and an indicative budget.
Share your team details, then choose a time on the calendar. No obligation to buy.