Developer Builds AI Receipt Scanner That Compares Grocery Prices Across Supermarkets
A developer created SmartBasket, an app that lets users upload a grocery receipt photo to find the cheapest prices across supermarkets, built as part of the 'Hack the Zero Stack with Vercel v0 and AWS Databases' Hackathon. The app uses EasyOCR and Groq's Llama 3 to extract and structure receipt data, while Amazon Aurora PostgreSQL handles product matching via fuzzy search using the pg_trgm extension. The architecture is decoupled across Next.js, FastAPI, Amazon S3, and SQS, allowing uploads to return instantly while AI processing runs in the background. SHA-256 hashing prevents duplicate receipts from being reprocessed, and Server-Sent Events deliver real-time updates to users once processing completes. The developer noted that PostgreSQL's built-in trigram search removed the need for a dedicated external search service, keeping the stack both cost-effective and scalable.
This is an AI-generated summary. ShortSingh links to the original source for the complete article.
Discussion (0)
Log in to join the discussion and vote.
Log in