Production AI/ML academy — isolated Python labs, pytest, FastAPI, MCP, optional Grok mentor
0
stars
53
commits
TypeScript
primary language
Aug 20, 2026
updated
Interactive training platform for modern AI/ML engineers: curriculum UI, architecture simulators, and evidence-backed Python labs.
server.ts) on port 3000XAI_API_KEY (optional; curriculum fallback without a key)labs/module-1 … labs/module-5 (isolated per-lab virtualenvs)Standalone static page in atlas/: who this is for, what is inside, what not to expect, and a filterable map of CPU-proved vs survey vs optional-live vs thin topics.
npm run dev # then http://localhost:3000/atlas/
Or open atlas/index.html without the applet. This is a shared learning-space pitch, not a paid product page. The in-app certificate remains a local keepsake.
cp .env.example .env # set XAI_API_KEY for live mentor replies
npm install
npm run dev # http://localhost:3000
Useful scripts:
| Command | Purpose |
|---|---|
npm run lint | Typecheck SPA + server.ts |
npm test | Vitest unit tests (certificate, quiz lock, progress parse) |
npm run build | Production Vite + server bundle |
npm run dev | Express + Vite middleware (development) |
Each lab has its own requirements.txt. Use a dedicated venv (dependencies conflict across labs):
cd labs/module-1-foundations
python3 -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt
pytest -q
See each lab README.md for validation/evidence commands. Cloud-agent notes live in AGENTS.md.
Stored in browser localStorage. Requires:
Private training applet. The in-app certificate is a learning keepsake, not a vendor-accredited credential.
TypeScript
54.3%
Python
42.2%
HTML
1.3%
CSS
1.0%
Production AI/ML academy — isolated Python labs, pytest, FastAPI, MCP, optional Grok mentor
0
stars
53
commits
TypeScript
primary language
Aug 20, 2026
updated
Interactive training platform for modern AI/ML engineers: curriculum UI, architecture simulators, and evidence-backed Python labs.
server.ts) on port 3000XAI_API_KEY (optional; curriculum fallback without a key)labs/module-1 … labs/module-5 (isolated per-lab virtualenvs)Standalone static page in atlas/: who this is for, what is inside, what not to expect, and a filterable map of CPU-proved vs survey vs optional-live vs thin topics.
npm run dev # then http://localhost:3000/atlas/
Or open atlas/index.html without the applet. This is a shared learning-space pitch, not a paid product page. The in-app certificate remains a local keepsake.
cp .env.example .env # set XAI_API_KEY for live mentor replies
npm install
npm run dev # http://localhost:3000
Useful scripts:
| Command | Purpose |
|---|---|
npm run lint | Typecheck SPA + server.ts |
npm test | Vitest unit tests (certificate, quiz lock, progress parse) |
npm run build | Production Vite + server bundle |
npm run dev | Express + Vite middleware (development) |
Each lab has its own requirements.txt. Use a dedicated venv (dependencies conflict across labs):
cd labs/module-1-foundations
python3 -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt
pytest -q
See each lab README.md for validation/evidence commands. Cloud-agent notes live in AGENTS.md.
Stored in browser localStorage. Requires:
Private training applet. The in-app certificate is a learning keepsake, not a vendor-accredited credential.
TypeScript
54.3%
Python
42.2%
HTML
1.3%
CSS
1.0%