Tender Radar: Finding Government Tenders for Small Contractors Locally

Tender Radar is a new open-source tool designed to help small contractors identify and manage government tenders directly from their own laptops. Built during the Hacktoberfest Open-Source AI Challenge, the application addresses the challenges faced by MSMEs in navigating complex tender requirements. By leveraging local-first AI, specifically Gemma 4 via Ollama, the tool extracts eligibility clauses, matches them against a firm's internal documents, and flags potential risks. It also utilizes TabPFN for win-probability estimation. A key feature is the 'citation guard,' which ensures that all AI-generated compliance statements are backed by verifiable evidence from the user's own files. Because the software runs entirely offline, it protects sensitive financial and operational data while eliminating the costs associated with third-party agents. This project demonstrates how small-scale, local AI models can provide significant operational advantages for businesses without requiring expensive cloud infrastructure or compromising data privacy.