Education / Nonprofit
EEF Learn
Open-source learning maps for the Emmanuel Eze Foundation — name a topic and a course agent builds a path you can walk, with every source credited.

EEF Learn

The challenge
The Emmanuel Eze Foundation went from feeding and teaching 100 children in Makoko to eight free bootcamps. Its next step was a place anyone can learn, not only children.
The people it had in mind had tried to teach themselves a skill and stalled. They had YouTube, blogs, and AI chat, but no structure, no sense of where they were, and nobody to talk to. And it had to run on a nonprofit budget.
Our solution
We built EEF Learn around a learning map: a topic becomes modules and lessons on a map you can move through in any direction, with finished work visible on the map.
Lessons hold readings, YouTube, video, audio, documents, code, quizzes, and flashcards, and every source is credited. A course agent researches the topic and fills the map while you watch, or a creator builds it by hand in the studio, then publishes it for anyone to travel.
Owlbert, the tutor, answers alongside the lesson. The foundation holds no model key: creators connect their own OpenRouter, OpenAI, Anthropic, or Google key, or their own assistant over MCP.
Technology used
- Next.js
- FastAPI
- PostgreSQL + pgvector
- Better Auth
- Turborepo
- Vercel
- Google Cloud Run
- Neon
The results
Live at emmanuelezefoundation.org/learn, alongside the foundation's own site.
Open source under the MIT licence and self-hostable with Docker, so the foundation's work is a public good.
Anyone 13 or older can preview any map for free; entering one takes a free account.
More case studies
Client OS

Agency Operations / Client Portal
Client OS
Unseen Engine's own app for running client work — every prospect and client on one hub, from the first website request to approval, with agents that draft and a person who releases.
Read the case studyOrder Intelligence

Hospitality / Event Catering
Order Intelligence
An AI order engine for multi-city coffee catering — ops answers a quiz about how they actually serve, then the system calculates quantities and lets a manager review, apply, or revert every suggestion.
Read the case study

Start a similar project
Let's discuss how we can deliver similar results for your business.