AI-Powered Research
Intelligent document analysis using pgvector embeddings for semantic search across student's research materials.
10xStudent
A comprehensive suite for students to automate document synthesis and focus on creativity. Built with TurboRepo, Next.js, TanStack AI, and PostgreSQL with pgvector.
10xStudent
A comprehensive student productivity suite that automates document synthesis and helps students focus on creativity. Built with modern technologies including TanStack AI for AI-native features.
Project Overview
10xStudent helps students automate repetitive tasks like note-taking, document organization, and research synthesis—so they can focus on what matters: learning and creating.
Key Technologies
- Framework: Next.js 15 with App Router
- Monorepo: Turborepo
- AI: TanStack AI with streaming support
- Database: PostgreSQL with pgvector for embeddings
- Auth: Clerk authentication
- Styling: Tailwind CSS
Features
1. AI-Native Architecture
Built with TanStack AI for reliable AI feature integration:
- Streaming responses
- Tool calling
- Token tracking and cost management
2. Vector Search
Using pgvector for semantic search across documents:
- Efficient embedding storage
- Similarity search
- RAG (Retrieval Augmented Generation) capabilities
3. Monorepo Structure
Shared packages for:
@repo/ui: React component library@repo/typescript-config: TypeScript configurations@repo/eslint-config: ESLint configurations
AI Workflow
This project showcases my AI-assisted development approach:
- AI agents for scaffolding boilerplate
- TanStack AI for building reliable AI features
- Proper error handling and streaming UX