AI Tool Recommender

01The Problem

The AI tools landscape has exploded to thousands of products across hundreds of categories. Searching for the right tool for a specific use case required hours of manual research across scattered review sites with no semantic understanding of user intent.

02The Solution

Built a Flask REST API with FAISS vector search and OpenAI embeddings over a scraped dataset of 4,000+ AI tools. LLM-generated summaries enable natural language queries that return highly relevant ranked recommendations.

03Outcomes

• ›Reduced user research time by 80% via semantic vector-based search
• ›Scraped and structured metadata for 4,000+ AI tools programmatically
• ›LLM-generated summaries enable intelligent, context-aware recommendations
• ›Built end-to-end from data pipeline to deployed REST API