Hacker
Manish Reddy
@manishr07
ContractSync — Cross-Stack API Contract Coordinator
Flip the Switch HackathonNov 2026
ContractSync is an AI-powered API contract coordination tool designed to prevent inconsistencies between backend and frontend development teams. When backend developers modify API responses, frontend TypeScript interfaces, mock data and API specifications can become outdated, leading to integration errors and broken applications. ContractSync introduces a collaborative workflow with two specialized engineering agents: • SpecArchitect analyzes OpenAPI specifications, identifies breaking changes, highlights compatibility risks and proposes migration strategies. • FrontendSync synchronizes TypeScript interfaces and mock data based on the approved API contract, then validates the changes using the TypeScript compiler. A human approval gate ensures that proposed API modifications are reviewed before implementation. The project includes an API contract analyzer, frontend synchronization engine, automated validation and an interactive workflow demonstration. ContractSync connects two specialized Google Antigravity agents to a shared Mattermost channel through Switch. SpecArchitect analyzes API changes and enforces human approval before modifying the OpenAPI contract. FrontendSync then synchronizes TypeScript interfaces and mock data, runs compiler verification and reports the results in the same channel. The workflow has been executed using the local Switch Core, Mattermost and Antigravity ACP infrastructure.
LaunchLens — Explainable Release Readiness with FalkorDB
Graph HacksOct 2026
LaunchLens is a graph-powered release-readiness agent that helps software teams identify hidden blockers before shipping. Using FalkorDB, it traverses task dependencies, traces upstream prerequisites, identifies downstream impact, and retrieves related decisions to explain why a release is or is not ready. Unlike a conventional task board, LaunchLens reasons over connected project data to show what gets unblocked by completing a task. Built with Python and Streamlit, it demonstrates explainable, graph-grounded decision support using synthetic project data.
Tankwise — Rain-Aware Water Tanker Planning
Environmental HacksOct 2026
https://builder.aws.com/content/3JarB6o9iQCKDfid1lN8JK5MGtw/tankwise-building-a-rain-aware-water-planning-tool-with-aws-serverless
ReachAi
First CommitSep 2026
https://builder.aws.com/content/3Jad7x2olCxg3Zm4udnbHFhOnvS/i-built-reachai-an-evidence-driven-ai-sales-assistant-with-strands-and-amazon-bedrock