Globally distributed MCP servers on WebAssembly at the edge
Build MCP servers as WebAssembly components with the wasmcp CLI and Spin, compose them into full servers, and deploy them to a globally distributed serverless platform with zero cold starts, so tools respond fast for users anywhere in the world.
online BOTH AI Agents & Automation
From AI Engineer — Zero Cold Starts: Serverless AI Agents on Akamai Functions — Thorsten Hans at 06:25
Problem: Traditional region-bound deployments give high latency to distant users, containers and VMs can cold-start for seconds to minutes, and hyperscaler platforms lock you into proprietary SDKs that are costly to migrate away from. WebAssembly components are tiny, portable and sandboxed, cold-starting in under half a millisecond.
For: Developers and teams exposing custom tools, resources and prompts to LLM clients (IDEs, agents) with a global user base and strict latency requirements.
Products from this video
Akamai Functions Linode MCP Inspector Ollama SpiderMonkey Spin Spin key-value store Vercel AI SDK wasmcp Zed
Examples
- A hello-world MCP server scaffolded in TypeScript with a single example tool, tested with MCP Inspector.
Other takes on AI agent platforms and compute access
- Autonomous AI agents that perform administrative labor for underserved small businesses, starting with dental practices
- A personal, open-source AI agent that can be operated through messaging channels such as WhatsApp and Discord
- Enterprise AI concierge for customer, sales, and operational workflows
- AI appointment-scheduling assistant for healthcare providers
- Autonomous SEC-filing research desk with self-graded scorecards
- Production-ready inference hosting for open AI models