At Flojoy, I have been involved with several key projects:
Stay tuned to learn more about how the Zeroth team and I built a comprehensive integration between Mecademic Industrial Robotics and Flojoy visual programming, including the design and thought process behind how a visual programming language should deal with robotics hardware.
During a walk in Seattle, I stumbled upon Artly, whose robot barista absolutely blew me away. When I got the opportunity to work with Flojoy, and I learned about the need for an demo that showcases Flojoy's potential and ease of use, the robot at Artly immediately came to mind.
The "coffee demo" involves the Mecademic Industrial robot executing a nice pourover, programmed entirely in Flojoy.
The idea of the demo was that Flojoy is so intuitive, it's easier to get the robot to make a pourover than to learn the technique yourself.
My conception of Flojoy -- visual functional programming.
My biggest contribution to Flojoy was the proposal and design for the Reactive Backend -- a project which turns Flojoy's previously run-once flowchart into a live, reactive control panel for any source of data, a lot like LabView.
Previously, Flojoy computed an execution graph and executed code on the completion of previous dependencies, but only once, and to watch for changes the entire flowchart needed to be completely re-run.
The reactive design changes this, instead creating a graph of observables (subjects in RxPY terminology) and using them to automatically re-compute parts of the flowchart as dependencies change in real-time.
This reactive rewrite also allows for Flojoy's robotics functionality.