VentureIQ is an AI-powered startup idea validation platform designed to give founders instant, comprehensive feasibility analysis. By leveraging a panel of specialized AI agents built with LangGraph, it assesses ideas from multiple perspectives—market size, technical feasibility, competitive landscape, and financial viability.

Features

• Multi-Agent Orchestration — Uses LangGraph to coordinate a team of specialized AI agents that debate and analyze startup ideas.
• Instant Feasibility Reports — Generates comprehensive validation reports covering technical, market, and business model risks.
• Cross-Platform Mobile App — A beautiful, intuitive Flutter app that lets founders capture ideas on the go.
• Robust AI Backend — Powered by FastAPI, Google Gemini, and OpenRouter for high-performance LLM routing.
• Vector Search — Integrates ChromaDB for semantic search and Retrieval-Augmented Generation (RAG).
• Scalable Infrastructure — Fully containerized with Docker, backed by PostgreSQL and Redis for caching and rate-limiting.

Tech Stack

• Mobile: Flutter (Dart) with Riverpod for state management and GoRouter for navigation
• Backend: FastAPI (Python 3.13) with async support
• AI/ML: LangGraph, Google Gemini, OpenRouter LLMs, ChromaDB
• Database & Caching: PostgreSQL 17, Redis 7
• Auth: Firebase Authentication with JWT token exchange
• DevOps: Docker, Docker Compose, Alembic (migrations), uv (Python package manager)

Architecture Highlights

VentureIQ separates concerns across a clean monorepo architecture:

The Mobile App
Built with Flutter, the client follows a feature-first architecture. It uses Riverpod for predictable state management and Dio for robust networking, ensuring a smooth experience even when agent processing takes longer.

The AI Backend
The core of the platform is a FastAPI service that orchestrates complex LLM workflows. It uses LangGraph to model the validation process as a state machine where different AI "personas" (e.g., Technical Lead, Market Analyst, Financial Advisor) interact. The backend relies on ChromaDB for vector storage to retrieve relevant market data, and Redis for rate limiting and fast response caching.

Infrastructure
The entire stack is containerized for local development and production deployment, using Docker Compose to spin up the API, PostgreSQL, Redis, and ChromaDB seamlessly.

Development Workflow

The project utilizes modern Python tooling, specifically uv for lightning-fast dependency management and ruff for linting and formatting. The database schema is managed through SQLAlchemy models and Alembic migrations, ensuring data integrity as the application evolves.