AI is creating a new kind of workforce bottleneck: the electrical, power, and cooling technicians who keep AI data centers safe, online, and scalable. GridForge gives aspiring critical-facilities workers a free place to practice the judgment those roles demand before they ever enter a live facility.
Instead of a chatbot or a static course, GridForge is a keyless interactive simulator. Learners begin with a plain-English primer, then work through branching job-site scenarios across data-center electrical systems, power and UPS systems, and liquid cooling. Each decision reveals a realistic consequence and a seasoned mentor’s explanation. Incorrect choices are explicit and safe: learners must choose the recommended response before advancing, so the product teaches the operational judgment behind safety and uptime rather than simply scoring a guess.
GPT-5.6 authored the 15-lesson training library at build time: the mentor voice, scenarios, branching outcomes, technical teaching, debriefs, and safety framing. Codex built the React simulator, reusable scenario engine, responsive product experience, and native animated SVG diagrams. The resulting app ships all learning content as static TypeScript data. It requires no account, API key, backend, or runtime AI/data call, so it is free, reliable, and usable offline after the first visit.
To try it, open the live app, choose Learn the basics for Lesson 1, complete the short primer, and start the GPU-rack cooling alert lesson. The README includes a judge quick-start, a precise keyless architecture explanation, and the project build log.
Built With
- codex
- gpt-5.6
- native-svg
- react
- typescript
- vercel
- vite
Log in or sign up for Devpost to join the conversation.