Exploring how AI and decentralized architecture could close the gap between what enterprise EAM systems promise and what plants actually get in year one.
01 — The problem
Twelve to twenty-four months to go live is typical. By the time the system is configured, the business case that justified it has aged out and the maintenance team has already built its own workarounds.
Field observations, drone inspection footage, thermography photos, voice notes from a technician on a scaffold — almost none of it reaches the CMMS in a structured, usable form.
When history is incomplete or wrong, planners stop believing it. Decisions move to spreadsheets and hallway conversations, and the asset register drifts further from reality every year.
02 — What we're exploring
We're looking at three things: faster ingestion of field data — voice, photo, sensor — so what a technician sees actually becomes a record; maintenance history that is verifiable and portable rather than locked inside one vendor's database; and predictive models trained on the real failure data most EAM systems never capture in the first place.
To be clear: this is not an attempt to replace SAP PM or IBM Maximo. Those remain the systems of record. The question we're asking is whether the layer around them — where the field data is lost today — can be rebuilt so the systems you already paid for finally get fed properly.
03 — Founder
Electrical Engineer · Dubai, UAE
23+ years in Oil & Gas maintenance and reliability, most recently as Chief Electrical Engineer and Head of Maintenance. Managed a $96M maintenance contract, built SAP PM programs from scratch, and ran RCM, RCA and RBI reliability programs on live operating assets.
04 — Talk to us
We're not selling anything — just trying to validate a real pain point before building. 15–20 minutes of your time to share what's actually broken.