ESP32-S3 Bench Guide
ESP32-S3 N16R8 bench guide: live pinout, lessons, simulators and the bench log.
Practise the IGCSE French speaking exam out loud. Get clear feedback on every answer and sit mock exams marked against the Cambridge 0520 speaking criteria.
EducationTypeScript · HTMLGitHub @RXcodeWalkerOnlineOwner unconfirmed
Have you tried it? One tap is enough. No sign-in needed.
No updates detected yet. GitHub releases will appear here.
A local-first web app for practising spoken French (IGCSE/A‑Level focus). It pairs a Vite + React frontend with a FastAPI backend for AI-assisted feedback, an offline regex-based coaching engine, and a gamified progression system.
Loading the full README.
Read the full README ↗A French speaking-practice app for IGCSE/A-Level learners. Users record spoken French, get AI feedback, earn XP, and track skill mastery.
This app spans three separately-deployed pieces — see docs/systems/topology.md for the full
map, deploy wiring, and environment variables.
src/ (this repo) — the Vite + React + TypeScript frontend. Deployed to Vercel.server/ (this repo, separate deploy) — a Node service hosting the Cambridge IGCSE French
(0520) scoring API. Deployed to Render (render.yaml, french-scoring service).backend/ — a separate git repository (its own remote and .gitignore) running
FastAPI, handling general coaching feedback, transcription, and exam endpoints. Deployed
separately to Render, dashboard-managed. See backend/README.md for its own setup.npm install
npm run dev # Vite dev server, http://localhost:5173
The frontend alone will run, but AI feedback and exam scoring need the other two surfaces
running too (or their env vars pointed at already-deployed instances) — see .env.example at
the repo root and docs/systems/topology.md.
For the FastAPI backend, see backend/README.md in that repository.
npm run dev # Vite dev server
npm run build # production build
npm run typecheck # tsc --noEmit, src/ only
npm run typecheck:scripts # tsc --noEmit, scripts/
npm run typecheck:server # tsc --noEmit, server/
npm run lint # ESLint
npm test # vitest run
Full command list, what each does and doesn't cover, and the Assessment-Engine change procedure:
docs/guides/development.md.
CLAUDE.md — entry point for AI coding agents working in this repo.docs/README.md — the documentation map and authority model.docs/systems/ — how the app fits together and what it must do (topology, the Assessment
Engine, the Supabase data model, the shop economy, adaptive difficulty).docs/guides/ — how to perform recurring tasks (content authoring, corpus coverage, the dev
workflow).docs/decisions/ — ADRs recording why the system is built the way it is.Not yet licensed for reuse.