Hands-on Codex training so your engineers get reliable, reviewed results from OpenAI’s coding agent in their own codebase.
10,000+ people trained across 500+ organisations
Codex is OpenAI’s coding agent. It plans and carries out work across a whole project, locally or in the cloud, and reviews code as well as writing it. If your firm already runs on ChatGPT, it is likely the coding agent your engineers will reach for first.
This half-day course follows the same structure as our Claude Code class, built around your stack, repositories and tickets. Each engineer takes a small, real build from brief to reviewed, tested change, with the guardrails that keep code, secrets and production safe.
What engineers said after our agentic coding class
100%
said the class will help them get more value from Claude Code at work
94%
enjoyed the class
79%
would recommend it to a colleague
5 hrs
average time each engineer expects to save every week
From our Claude Code class, which this course mirrors. Post-class surveys from the engineering cohorts at a 400-person UK consumer technology company, July 2026.
Every session is built from scratch around your work, but every session builds the same foundations.
Setting up Codex in the app, terminal or editor, choosing how much it may do before asking for approval, and writing an AGENTS.md that records your conventions, tooling and access.
Scanning repos, tickets, logs and errors for opportunities with evidence behind them, then committing to one small build for the day.
Extracting your coding style from the repo with file references, saving it to AGENTS.md and using it to bring new code into line.
Having Codex review a vague ticket for ambiguity and edge cases, turning the answers into a one-page brief, then planning before any edit with human approval.
Choosing the model and reasoning effort with cost in mind, keeping context clean between tasks and steering the work as it goes.
Running independent tasks side by side and handing longer jobs to Codex in the cloud, then reviewing the result before anything merges.
Using Codex’s built-in code review alongside a sceptical review prompt, debugging one hypothesis at a time, and hardening risky code for injection, validation and load.
A verification plan with unit and end-to-end tests, browser QA that tries to break the app, and a PR summary of user impact, risky paths, tests and rollback.
The capstone
Every engineer takes one real build from their own repo all the way to a reviewed, tested change, so the team leaves with shipped work and a repeatable way of working, not just notes.
We can usually map your highest-impact use cases in a single conversation.
Get in TouchThese came out of our agentic coding classes. Each one is a habit that transfers directly to Codex.
A sceptical review of a vague feature ticket that surfaces ambiguities, unhandled edge cases and likely misreadings, then turns the answers into a sharp implementation brief.
An agent clicks through member-facing flows like a real user, tries to break them, reproduces failures and traces the cause in code before proposing a fix.
Turns a finished change into a review-ready summary of user impact, risky paths, tests and rollback, and flags anything unrelated that slipped in.
Fixes injection risks, N+1 queries and missing error handling in a risky endpoint, explaining each fix and showing the diff before anything is accepted.
Drawn from our anonymised client use case library.
Three signals showing how fast agentic coding is becoming standard practice. Each links to its source.
The day runs in four parts. Each engineer works on their own small build throughout, so every exercise moves real work forward.
01
Setup and approval modes. How LLMs generate code and why deterministic tests matter. A read-only first look at your tooling. Finding something worth building and capturing house style in AGENTS.md.
02
An ambiguity review of the ticket and a sharp one-page brief. Planning first with human approval. Model and reasoning choices with cost in mind. Steering live. Parallel and cloud tasks.
03
Built-in code review plus a sceptical reviewer prompt. Debugging one hypothesis at a time. A hardening pass for injection, validation, queries and error handling.
04
A verification plan with unit and end-to-end tests. Browser QA that tries to break the app. Guardrails for secrets and least privilege. A PR summary covering impact, risk and rollback.
Delivered in person at your offices, or remotely.
Engineering leads rightly ask what happens to their code. Safety is built into the day rather than bolted on at the end.
Engineers start with Codex asking for approval before it changes anything, and loosen that only for tasks they trust
The first look at your tooling is read-only, and review prompts never modify code
No .env files opened, no credentials printed and no live endpoints called during exercises
Secrets, least privilege and review before every merge are taught as a dedicated guardrail
Work done in the cloud is reviewed before anything merges
Anyone without a repo they can safely use works on our practice repository instead, so no client code is ever required
Going further
Once a team has the habits, the next step is making them a standard across every repo and engineer.
Teams whose firm runs on ChatGPT and want to use Codex properly rather than by trial and error.
Developers working on real products and integrations who need changes that are tested, reviewed and safe to ship.
Leads and platform teams deciding conventions, guardrails and how agentic coding fits the review process.
Data engineers and BI developers who want the same rigour on pipelines, scripts and reporting code.
Other courses teams pair with this one.
Every Codex session is led by one of our instructors, with a dedicated tutor in the room so nobody gets left behind.
Founder + Chief AI Officer
AI Instructor
AI Instructor
AI Instructor
AI Instructor
AI Instructor
AI Instructor
AI Instructor
Yes. Each engineer works in a repository they can safely use, so the brief, the build and the review are all real. Anyone without one uses our practice repository instead.
Yes. The syllabus covers AGENTS.md, approval modes, planning before building, model and reasoning effort, parallel and cloud tasks, Codex’s built-in code review, hardening and testing AI-written code.
Engineers start with Codex asking for approval before it changes anything, first contact with your tooling is read-only, exercises never open .env files or call live endpoints, and secrets, least privilege and review before merge are taught as a dedicated guardrail.
Usually whichever your firm already pays for. The habits we teach carry across both, and we run the same day on Claude Code for teams that use Claude.
Yes. Our Codex training for non-developers teaches analysts, finance and operations teams to build their own tools from plain English. Many firms run both.
In person at your offices by default, with an instructor and a dedicated tutor in the room. We can run it remotely if that suits your team better.
Yes. We are based in Covent Garden, London, and run most sessions in person at clients’ offices in London and across the UK. We also work with teams worldwide, from Paris and Berlin to New York and Singapore, and can run sessions remotely when that suits better.
From The Distillation, our newsletter on putting AI to work.
GPT-6 Astra and Claude Opus 5.5 moved the frontier twice in three weeks. Plus AI training for the industries that make things, field notes from our US rollouts with Access Holdings portfolio companies, and funded graduate places on our next open enrolment sessions.
Josh Page ·
GPT-5.4 Thinking is a meaningful accuracy improvement, but the bigger story is what it says about where practical AI adoption is heading: better models, reusable workflows, and more interest in agentic ways of working.
General Purpose ·
Reasoning models do more than predict the next token. They break problems into steps, use tools, inspect evidence, and adapt their approach as they work. That makes them especially useful for analytical and multi-step business tasks.
General Purpose ·
Tell us about your stack and how your engineers work, and we will design a day around your repositories and the way your team ships.
Get in Touch