How aishwarya0903/adaptive-question-generator-using-candidate-response-patterns works
AdaptIQ is an adaptive interview engine: an interviewer starts a role-specific session, the candidate answers questions, and the backend scores each response before selecting a graph branch or generating an LLM follow-up. At session end, the engine summarizes performance and provides an AI recommendation; the dashboard presents question-effectiveness and candidate-trend analytics. The graph shows the sampled API, scoring, graph, and UI relationships plus README-documented capabilities. Authentication is listed in the README’s tech stack, but no implementation is evidenced in the supplied source.
Main entry points: main.py Interview and analytics API · llm.py Prompt chaining · Interview.jsx Interview flow · App.jsx App navigation
Components 15
Interview UI
- Interview flow
frontend/src/Interview.jsx - App navigation
frontend/src/App.jsx - Frontend entry
frontend/src/main.jsx
Interview runtime
- Interview and analytics API
backend/main.py - Session state
backend/main.py - API data models
backend/models.py - Response scorer
backend/scorer.py - Prompt chaining
backend/llm.py
Question graph
- Question graph operations
backend/graph.py - Question graph seed
backend/seed.py
Analytics
- Analytics dashboard
frontend/src/Dashboard.jsx
Other components
- Interviewer
- Candidate
- Neo4j
- Claude API
Connections 18
- Interviewer → Interview flow: starts session
README.md - Candidate → Interview flow: submits answers
README.md - App navigation → Interview flow: renders
frontend/src/App.jsx - App navigation → Analytics dashboard: renders
frontend/src/App.jsx - Frontend entry → App navigation: renders
frontend/src/main.jsx - Interview flow → Interview and analytics API: calls endpoints
frontend/src/Interview.jsx - Analytics dashboard → Interview and analytics API: fetches analytics
frontend/src/Dashboard.jsx - Interview and analytics API → API data models: uses contracts
backend/main.py - Interview and analytics API → Session state: reads and updates
backend/main.py - Interview and analytics API → Question graph operations: fetches and saves
backend/main.py - Interview and analytics API → Response scorer: scores answers
backend/main.py - Interview and analytics API → Prompt chaining: generates follow-ups
backend/main.py - Interview and analytics API → Prompt chaining: generates recommendations
backend/main.py - Prompt chaining → API data models: uses score type
backend/llm.py - Response scorer → API data models: returns score
backend/scorer.py - Question graph operations → Neo4j: queries and writes
backend/graph.py - Question graph seed → Question graph operations: seeds questions
backend/seed.py - Prompt chaining → Claude API: requests completions
backend/llm.py