Quick Question

Inspiration

We created QuickQ to empower tech job seekers by enhancing interview preparation, inspired by the competitive tech job market and the AI in Action Hackathon’s focus on Google Cloud and MongoDB.

What it does

QuickQ is a full-stack app with a Flask API and an Android application:

AI Job Search: Uses Google Cloud’s Vertex AI and MongoDB vector search to match 9,300+ job postings.
Interview Questions: Generates tailored technical questions from a 200+ question bank.
AI Feedback: Provides Vertex AI-driven feedback on practice answers.

How we built it

Backend: Flask API with Vertex AI for NLP and MongoDB Atlas for job/question storage. Hosted on Google Cloud with Docker.
Frontend: Native Android app using Compose for UI, Retrofit for networking, Room for caching all written in MVVM clean architecture.
Process: Sourced dataset, built API endpoints (/jobs, /questions, /feedback), and designed an intuitive frontend.

Challenges we ran into

Cleaning the dataset for MongoDB.
Tuning Gemini prompts for accurate, low-latency responses.
Syncing frontend-backend for real-time features.
Balancing features within hackathon timeline.

Accomplishments that we're proud of

Integrated Vertex AI and MongoDB for scalable AI features.
Delivered a full-stack MVP with polished UI.
Released open-source on GitHub.

What we learned

Mastered Vertex AI and MongoDB vector search.
Improved full-stack API and UI design.
Enhanced problem-solving under time constraints.

What's next for QuickQ

Enhance MongoDB integration to improve search performance and derive greater value from the datasets.
Enhance semantic search and feedback with advanced AI.
Implement analytics for user performance tracking.