AI engineering agency · London & remote
AI agents,
built to work.
We design, build and run production AI agents that take real work off your team's plate. Then we help you build the team and processes to own them.
ops-agent · example
OnlineReconcile last week's invoices against open POs.
- erp.list_invoices(week=42)
- erp.match_po(×37)
Matched 35 of 37. Two need a human: a price mismatch and a missing PO. Sent to finance for approval.
Model-agnostic
Claude, GPT, Gemini or open-weight. Chosen by evals on your task, swappable later.
Agency
We build AI agents for your team.
Academy
We teach engineers to build them.
What the agency does.
What we do
Five things, done properly.
Example engagement
All worksupport-agent · queue
- Refund request · duplicate chargeResolved
- Change delivery addressResolved
- Chargeback disputeEscalated to human
A support agent that resolves routine tickets and hands the rest to a person, with full context.
How we work
EngagementsWk 1–2
Scope
Workflow, data, tools and success metrics.
Wk 3–4
Prototype
A working agent on real data, evals from day one.
Wk 5–10
Harden
Integrations, guardrails, tracing, rollout.
Ongoing
Operate
Monitor, improve, hand over or stay on.
What the Academy teaches.
- Path 1Start here
LLM and Agent Foundations
3 courses · ~56h
- · LLMs for Engineers
- · Agent Foundations
- · Building Agents with Claude and eve
- Path 2
Evaluation and Optimisation
5 courses · ~135h
- · Evaluating and Scoring Agents
- · Agent Benchmarks and Simulations
- · Prompt and Context Optimisation
- · Human Feedback at Scale
- · Fine-Tuning and RL for Agents
- Path 3
Documents, Retrieval and Memory
3 courses · ~60h
- · Document Ingestion Pipelines
- · Retrieval and Knowledge Systems
- · Agent Memory
- Path 4
Production AI at Scale
4 courses · ~72h
- · Infrastructure for Agents
- · Production Agent Systems
- · Scaling AI Systems
- · LLM Cost Engineering
- BeginnerAI agents · Claude Code
Agent Foundations
What an AI agent really is, how it uses tools and context, and how to make one dependable enough for high-stakes work.
6 lessons ~18hFree - IntermediateAI agents · Evals
Evaluating and Scoring Agents
Turn "it seems better" into numbers you can defend: error analysis, code graders, calibrated LLM judges and honest statistics.
7 lessons ~20hFree - IntermediateAI · Transformers
LLMs for Engineers
How large language models are built, trained and run, so you can reason about their behaviour, cost and failure modes in production.
9 lessons ~18hFree Setup guides
- Install Claude Code
- MCP servers in Claude Code
- Get an OpenRouter key
- Durable agents with eve
2019
Founded in London
37
Cities our community reached
20
Countries
5
Practice areas, one team
About
Engineers who build agents. Not slides.
Our story2019
Coders In Hoods starts in London as a coding community and course.Growth
The community grows to 37 cities in 20 countries.Crew
Engineers from the community start taking on client projects together.Focus
The work centres on AI agents, evals and the teams around them.Today
An AI engineering agency, building agents in production.
- 01
Senior hands on keyboards
The people you first talk to write your code. No bait and switch.
- 02
Evals before vibes
Every agent ships with a test suite. We report numbers, not anecdotes.
- 03
Boring is a feature
We pick the simplest architecture that works, and say no to hype.
- 04
You own everything
Code, prompts, evals and infra stay in your repos and your cloud.
Give an agent the boring part.
Tell us what you're working on. A senior engineer replies within one business day.
Start learning, free.
Every course is free. Sign in to start, and your progress, quiz scores and bookmarks follow you.