Scaling Your AI Agents with Open Source: API ➔ No-Code MCP & Observability wit…
summary
Deploys an open-source, no-code proxy called reShapr that converts OpenAPI/GraphQL/gRPC specs into MCP tools on the fly, without backend code changes. Claims a 10x reduction in MCP tools for a GitHub GraphQL use case (200+ → 1) and lower latency/token use by collapsing and renaming operations, filtering inputs/outputs, and securing credentials. Shows CLI/UI demos, Kubernetes charts, observability hooks and full support for the Model Context Protocol (MCP).
AI-written summary · about
chapters (12)
- 0:00Introduction
- 1:55Enterprise MCP challenges
- 4:26Introducing Reshaper
- 6:15Live demo setup
- 9:46Advanced GraphQL demo
- 13:38Custom tooling optimization
- 17:14Reshaper benefits
- 20:03Security readiness
- 22:09Kubernetes deployment
- 23:14Observability metrics
- 23:56MCP server comparison
- 24:59Call to action
related talks
36mPutting AI Agents to Work: A New Era of Open Source Connectivity with MCP
30mComposable AI: Building Next-Gen AI Agents with MCP
27mPanel: APIs for AIs
18mScaling AI Agents With gRPC: Intuit's Enterprise MCP Service Marketplace
50mBeyond local tools: Dive into the Model Context Protocol (MCP)
01h 30mBuilding for the agentic future
23mMCP = Mega Context Problem
46mMCP for LLM Agents, Infrastructure APIs & Data Services