Hey all — I’m Hexi from Bumblebee. We’re a MSP focused AI Workflow Builder taking on Rewst and N8N. We’re piggybacking on Hudu’s Build-a-Feature format to share an early integration concept and gather feedback.
Build-a-feature: Turn Processes into Workflows that actually take actions
Some processes are repetitive button clicks and tab switches that shouldn't even exist in the first place. If I could get my tools to talk to each other, I could automate many of them and save myself a vacation.
The main idea:
Use an AI assistant to turn documented processes into a workflow that can do the work. You already have done the hard work of defining and consolidating processes; the button clicks should be delegated to a workflow that can make those updates in each system. If AI is so good at writing code, it should be able to build those integrations for me as I need it, and assemble them into a workflow based on my process.
How it works:
In your process documentation, there is a button you can click to start a conversation with an AI assistant - it will read your documentation, map that to different API calls for different integrations, and build a visualizable workflow that can be carried out autonomously. If the processes are not automatable - (e.g. because no API is available, or too sensitive to delegate), the AI assistant should tell me upfront so I stop wasting my time.
After running the workflow, it brings up the result directly to you for a side-by-side comparison against your documentation. When you gain enough trust, you can set the workflow to run on an automated trigger.
A quick look:
Initial Use Cases:
User onboarding / offboarding process
License renewal reminder process
Ticket labor log refinement process
Initial Integrations:
PSA: HaloPSA, ConnectWise, AutoTask, Syncro
Documentation: Hudu, IT Glue
Workspace: Microsoft Graph, Google Admin, CIPP
Security: Avanan, etc.
CRM: HubSpot
Help us shape it:
Tell us if you'd be interested to try out this feature
Tell us about what manual process you'd like to automate the most