Jaison John Samuel J@Jaison

Enterprise procurement and security teams spend weeks manually reviewing software vendors to ensure compliance with data privacy laws and security certifications (such as SOC2, ISO 27001, and GDPR). This manual "Trust and Safety" bottleneck delays critical software adoption and exposes companies to massive risks if a hidden vulnerability is overlooked. VendorGuard AI is an AI-powered SaaS product and internal business tool built to completely replace this manual process. By combining real-time web scraping with frontier Large Language Models, VendorGuard AI acts as an agentic multi-step application, condensing a 6-week manual audit lifecycle into a 15-second autonomous workflow. How It Works: Autonomous Scraping (Bright Data MCP): Our agentic workflow safely bypasses bot protections to scrape a target vendor's official domain, privacy policies, and public security documentation in real-time, converting messy HTML into clean markdown. AI Reasoning (AI/ML API): The backend ingests the scraped markdown, utilizing strict JSON-schema prompting to extract compliance metrics, cross-reference breach databases, and calculate a weighted Trust Score. Evidence Explorer: To eliminate AI hallucinations, every compliance claim generated by the LLM includes a direct citation, a confidence score, and a link back to the exact source text on the vendor's website. Built entirely on Native.builder, VendorGuard AI features a premium, Series-A grade glassmorphic command center. It includes a live terminal-execution log, real-time risk gauges, interactive compliance checklists, and a frictionless local session manager that allows procurement officers to run audits instantly without database bottlenecks.

🌎 Online Hackathon | 📅 August 3–10, 2026| 💻 Powered by native.builder The official native.builder hackathon, hosted by NativelyAI and lablab.ai. Join founders, developers, designers, and builders to create real AI-native applications powered by native.builder: from idea to deployed product, fully in the cloud. 💻 This hackathon is fully online. No travel required.