Haskell Devs Deploy AI-Powered 'Greet' Tool to Cloud Run, Accidentally Prove Monads Are Useful

Google AI's New Year, New You Portfolio Challenge has led to the development of a Model Context Protocol (MCP) application using Haskell, Cloud Run, and Gemini CLI. The project demonstrates the use of Haskell, a functional programming language, to build an MCP server that can be deployed to Google Cloud Run. The MCP protocol allows for language-independent implementation, enabling developers to use the latest AI approaches with languages other than Python. The project involved setting up a Haskell development environment, building a minimal MCP server, and deploying it to Google Cloud Run using a Dockerfile and Cloud Build. The remote server was then validated using Gemini CLI, a command-line interface for interacting with MCP servers. The project showcases the potential for using Haskell in AI development and demonstrates the ease of deployment to cloud-based platforms like Google Cloud Run, highlighting the flexibility and scalability of the MCP protocol.