DEV Community

prasanna-lakshmi18
prasanna-lakshmi18

Posted on

AI-powered Study Companion for GATE Aspirants with Redis 8

#redischallenge #devchallenge #database #ai

Redis AI Challenge: Beyond the Cache

This is a submission for the Redis AI Challenge: Beyond the Cache.

What I Built

I built an AI-powered study companion tailored for GATE aspirants. The app goes beyond being a simple chatbot—it provides:

  • Real-time question answering from a curated syllabus-based knowledge base.
  • Audio-based interaction for hands-free learning, using streaming voice input and output.
  • Personalized recommendations for study topics based on previous performance.
  • Semantic search over vast study materials.
  • Daily quizzes with adaptive difficulty to target weak areas.

The goal was to create an engaging, accessible, and always-learning AI tutor that adapts to each student’s needs.

Demo

🎥 Video demo: here
🖥 Live app:App here

For more information visit my github repo:Repo

How I Used Redis 8

Redis was the real-time backbone of this app, not just a cache:

  1. Vector Search with Redis Stack

    • I stored embeddings of GATE syllabus topics, lecture notes, and past question papers using Redis Vector Search.
    • This enabled fast semantic search for relevant concepts when a user asked a question.

  2. Streams for Live Audio Interaction

    • Redis Streams handled live audio data from the user and sent it to the AI model in real-time.
    • This powered a streaming Q&A mode without lag.

  3. Pub/Sub for Real-time Quiz Updates

    • When a student takes a quiz, results are published instantly, and other connected sessions (like progress dashboards) update live.

  4. Hash & JSON Structures for User Profiles

    • RedisJSON stored student profiles, past performance metrics, and preferences.
    • This allowed the AI to dynamically tailor study recommendations.

  5. Time-series Data for Progress Tracking

    • Using RedisTimeSeries, I logged study sessions, quiz scores, and topic completion rates.
    • This powered analytics to help students visualize their improvement over time.

By combining Vector Search, Streams, Pub/Sub, JSON, and TimeSeries, I turned Redis 8 into the primary real-time data layer for the AI tutor—far beyond simple caching.

Top comments (0)