Product releases, notes from the field, and what we are working on. Longer posts link out to LinkedIn or X.
Process Alert generates notifications from inside AVEVA Application Server rather than polling it from outside, so alerts carry the equipment context the object model already holds: the setpoint, the current value, and the area the alarm came from. No per-tag configuration to maintain.
Recorded training and meetings are only useful if someone can find the one minute that answers their question. This assistant transcribes a video library, indexes it, and answers questions about it in plain English.
Every part of it runs on hardware you control: the speech-to-text engine, the search index and the language model. No video, audio or transcript is sent to a third-party AI service. We run one of these ourselves, and we build them for customers who want their own.
Tell us what you're working on: plant, platform, and what you're trying to achieve. We'll scope it and propose an approach.