Happy Friday, FME'ers! Welcome back to TGIF: Thank Goodness It's FME.
This week, we're doing something a little different. Instead of spotlighting a single topic, we asked our team one simple question: what's your favourite FME feature? The answers were passionate, practical, and honestly, a little nerdy (in the best way). This is Part 1 of a two-week series — and we're kicking things off with Zoe and Aaron, who both landed on features that are changing how FME connects with the rest of the world. Let's hear from them!
Zoe’s Favourite Feature
If you’ve ever wanted to share your FME workspace logic with the world or even just with another team, without anyone needing to know what FME is, Data Virtualization is Zoe’s answer.
Data Virtualization in FME Flow lets you publish any FME workspace as a custom REST API endpoint. No third-party API gateway, no extra code. You define the endpoint name, the parameters, and the workspace that powers it, and FME Flow handles the rest, including automatically generating interactive Swagger documentation so anyone can discover and call your API right away.
In their demo, Zoe walks through the full process from scratch: creating a new API in FME Flow, defining a custom endpoint, connecting it to a workspace in FME Workbench, and publishing it live. They then show the finished API in action, calling it from Postman and seeing the workspace results come back in real time. It is the kind of workflow that takes something complex like a workspace with real business logic, and makes it instantly accessible to anyone who can call an HTTP endpoint.
Check out Zoe’s demo here!
Aaron’s Favourite Feature: MCP Server Support on FME Flow
Aaron's pick is one for anyone excited about where AI and data integration are heading: MCP Server Support in FME Flow.
MCP, or Model Context Protocol, is an open standard that lets AI assistants connect to external tools and data sources. With FME Flow's built-in MCP Server support, you can turn any FME workspace into an AI-callable tool. Create the tool in FME Flow, point it at a workspace, and any MCP-compatible client can call it by name, in plain language.
In his demo, Aaron shows just how practical this gets. He walks through a setup where FME workspaces back tools for checking live traffic conditions near a location, monitoring SSL certificate expiry, and more. The AI client asks a question, FME runs the workspace, and the answer comes back in seconds. No manual steps, no extra code. It is the kind of integration that turns your existing FME logic into something an AI assistant can actually act on.
Check out Aaron's demo here!
That's a wrap on Part 1 of Team Favourites! From publishing your workspace logic as a custom REST API to putting your FME tools in the hands of an AI assistant, it's clear the future of FME is wide open.
Next week, we're back with Part 2 — Shirley, Carl, and Mac share the features they can't live without.
In the meantime, we’d love to know: what’s YOUR favourite FME feature? Drop it in the comments and we might just feature it in a future post. Happy FMEing, everyone!