Skip to content

SideQuest

An advanced AI-powered physical skill transmission and hands-free co-piloting ecosystem designed to digitize local vocational trades (heritage crafts, electrical repair, knot tying) and guide students in their native reg

Code Video Cerebral Valley

Core trick

Segments expert demo video into timed steps with validation checklists, then uses webcam vision to check a learner's attempts against them.

Tags

educationvision

Description

An advanced AI-powered physical skill transmission and hands-free co-piloting ecosystem designed to digitize local vocational trades (heritage crafts, electrical repair, knot tying) and guide students in their native regional Indian languages using real-time computer vision and voice commands.
nirranjan121/WOTF_AI★ 01 forksHTMLCreated Jul 10, 2026 · last push Jul 11, 2026

README excerpt

# Watch Once, Teach Forever (WOTF AI) — Kushal.ai 🏺

An advanced AI-powered physical skill transmission and real-time co-piloting ecosystem. WOTF AI is designed to digitize local trade skills (e.g., heritage crafts, electrical repair, knot tying, and basic sciences) and guide vocational learners in their native regional Indian languages.

---

## 🚀 Key Features

### 🎬 Phase 1: Expert Video Digitization & Synthesis
- **Webcam & File Upload**: Ingests expert performance footage or live demonstrations.
- **Dynamic Step Segmentation**: Chronologically segments continuous actions into logical milestones.
- **Validation Checklist Synthesis**: Generates strict visual checks with associated confidence thresholds.
- **Safety Hazard Identification**: Pinpoints potential errors, mapping out mitigation warnings and vocal guidance.
- **Serialized Protocol Schema**: Outputs validated profiles adhering strictly to robust [Pydantic](file:///Users/nirranjannaarayanmr/Downloads/Compiler%20part1/schema.py) structures.

### 🎯 Phase 2: Live Hands-Free Visual & Audio Coaching
- **Continuous Camera Stream**: Evaluates student performance frame-by-frame.
- **Dual-Model Arbitration**: 
  - **Fast-Loop (Gemma E2B)**: Executes instant visual confirmations and safety checks.
  - **Slow-Loop (Gemma E4B)**: Escalates near-threshold or ambiguous actions to deep reasoning layers to avoid false positives.
- **Interactive Step Visual Guide**: A lightweight, real-time client-side SVG drawing engine that illust

Team