SANYAMWADHWA.TECH :: PROJECT :: FLOWSYNC --- SUMMARY ---TITLE............. FlowSync — AI Project Memory for Coding AgentsDATE.............. March 2026CATEGORY.......... Agentic AIAWARD............. AI for Bharat HackathonPROBLEM........... AI coding agents are stateless. Every session starts from zero with no memory of past decisions or architecture. FlowSync gives agents persistent project memory via the Model Context Protocol.WHAT IT DOES...... Gives AI coding agents persistent memory across sessions via the Model Context Protocol. Agents log why they made decisions and query weeks of project history; git pushes auto-capture context as a fallback, all through a serverless AWS Bedrock pipeline. --- RESULT ---METRIC 1.......... ~1.2s median query latencyMETRIC 2.......... ~200ms ingestionMETRIC 3.......... ~$4 / dev / month --- WHAT I BUILT ---HIGHLIGHT 1....... Event-driven AWS pipeline: raw events land in DynamoDB in about 200ms, Bedrock extraction runs asynchronously, and branch-aware RAG answers queries at ~1.2s median latency.HIGHLIGHT 2....... Tiered model routing sends once-per-commit extraction to Nova Pro and interactive chat to Nova Lite, cutting inference cost by roughly 60%. --- HOW IT WORKS ---STEP 1............ VS Code Extension + MCP Server (TypeScript)STEP 2............ API GatewaySTEP 3............ AWS Lambda (Python 3.12)STEP 4............ Amazon Bedrock (Nova Pro + Nova Lite) + Titan EmbeddingsSTEP 5............ DynamoDB + S3STEP 6............ Next.js Team Dashboard --- FEATURES ---FEATURE 1......... 5 MCP tools discovered natively by Copilot, Cursor, and Claude CodeFEATURE 2......... log_context records the why behind each change: decisions, risks, and reasoning per commitFEATURE 3......... search_context answers natural-language questions with strict source citations back to commitsFEATURE 4......... Git-push auto-capture fallback, so the project brain grows even for developers without an agentFEATURE 5......... Titan embeddings stored beside context items in DynamoDB; no separate vector database to operateFEATURE 6......... Branch-aware RAG plus a real-time team dashboard of decisions, risks, and pending tasksFEATURE 7......... Shipped as a VS Code extension with scrypt-hashed project tokens and a 1-hour RAG response cache --- STACK ---STACK............. AWS Bedrock, Lambda, DynamoDB, RAG, TypeScript, Python, MCP --- LINKS ---CODE.............. https://github.com/SanyamWadhwa07/flowsyncHUMAN PAGE........ https://sanyamwadhwa.tech/projects/flowsync