Software that understands building controls
Mason builds AI tools for the people who program HVAC and building-automation systems. Our first product, the Mason AutoProgrammer, turns a plain-language description of equipment into a real Niagara wire sheet.
Controls programming is slow, repetitive work: the same sensors, the same PID loops, the same alarm logic, wired by hand on every job. The knowledge to do it well is real and specific — and most of it lives in engineers' heads and a stack of manufacturer documentation.
We think that knowledge belongs in a tool that can apply it. So we're building an AI that has read the controls library and can assemble a wire sheet from a sentence — then hand it back to the engineer to review, tune, and stand behind. The goal isn't to replace the engineer. It's to give them a faster start and fewer repetitive hours.
How we build
Grounded, not guessing
Our tools are built on real controls knowledge. The agent reaches for the right block because it knows the catalog — it isn't improvising.
Engineer in the loop
AI does the wiring; the engineer makes the calls. Everything we build stays inspectable and tunable, never a black box.
Take out the grunt work
Controls engineers should spend their time on sequences and judgment, not dragging blocks. We automate the tedious part.
We'd love to hear from controls teams
Whether you want early access or just want to talk shop about controls and AI, reach out.