InterVu researches your target company, builds a personalised knowledge base, then conducts an adaptive AI interview — scoring every answer in real time.
The Full Pipeline
Not generic questions. Real company data, real interview patterns, real feedback.
Scours Tavily + Firecrawl to pull real job descriptions, interview experiences, and engineering blogs.
Gemini extracts structured skills, topics, interview rounds, and difficulty from noisy web content.
Content is chunked, embedded with sentence-transformers, and stored in ChromaDB for semantic retrieval.
Generates a custom interview plan — sections, question counts, and focus areas tailored to the role.
LangGraph orchestrates a multi-turn interview that adjusts difficulty based on your real-time performance.
Post-interview report with scores, strong/weak topics, and personalised study recommendations.
How It Works
Tell InterVu where you're interviewing — e.g., NVIDIA PTX Compiler Intern or Google L3.
Agents search the web, scrape pages, and extract structured knowledge into ChromaDB.
A custom interview plan is created with adaptive sections and difficulty calibration.
Answer questions. The AI evaluates each answer and adjusts the next question accordingly.
Receive an overall score, topic breakdown, and personalised improvement recommendations.