Onboard is how building experts talk to buildings. The only self-serve AI platform for connecting and communicating with your buildings and its equipment.
Building Energy and Optimization Engineers use Onboardās AI software to acquire the building data they need to improve their clientās buildings. Onboardās self-serve AI software enables any engineer to collect, standardize, visualize, and share building data across their team or analytical apps. Engineers can flexibly install Onboardās virtualized software on their ITās preferred hardware inside the buildingās network. Currently, Onboard is focused on BACnet and Modbus data.
Onboardās AI filters out irrelevant data and applies standard ontologies, including Project Haystack Tags and Googleās Digital Buildings Ontology. Onboard transparently shares how its AI achieved its predictions, which allows engineers to intuitively operate its AI and all other software operations entirely by themselves.
Engineers can export their data to CSV, JSON or use Onboardās REST API to pipe data to any application. Onboard provides Python and R SDKs for engineers seeking to flex their data science skills as well as an ongoing YouTube Series and Knowledge Base for how to creatively use your building data.
Onboardās uniquely lightweight approach makes Onboard attractive to smart building consultants and OEMs alike, anyone needing a scalable, transparent and a cost-effective method for normalizing building data across portfolios. Onboard is an enabler of use-cases; its mission is to equip every engineer with the data they need to improve the worldās buildings.
Amazon's Niharika Kishore described how a model with one technician covering 50 buildings pushed the company toward AI-assisted maintenance for HVAC, refrigeration, and water. The point was not adding more alarms, but finding a way to act on them before breakdowns hit the site.
Hannah Baker, engineer at Willow, walks through how DFW Airport built a CBM program that actually stuck, from training a non-technical QA team to triage thousands of faults, to graduating recurring issues into automated work orders, to tracking a single KPI called 'unsuccessfully actioned' that finally gave leadership visibility into whether closed work orders were actually fixing the problem.
Jose de Castro, CTO of Mapped, shows how one of the world's largest retailers moved restroom operations from schedule-based janitorial rounds to condition-based workflows by combining foot traffic sensors, flush counts, soap levels, and occupancy predictions into AI-summarized work orders that land directly in the existing CMMS, with no new dashboards or tools for technicians to learn.
Brad Dameron from the University of Iowa's Asset Optimization Team and Katie Rossman from Clockworks Analytics walk through how Iowa handles 3,500 faults per day without burying their maintenance shops, showing the exact triage, routing, and closeout workflow they built to turn fault detection into planned work orders that look and feel identical to every other work order in the system.
Head over to Nexus Connect and see whatās new in the community. Donāt forget to check out the latest member-only events.
Go to Nexus ConnectJoin Nexus Pro and get full access including invite-only member gatherings, access to the community chatroom Nexus Connect, networking opportunities, and deep dive essays.
Sign Up
Nexus Labs is the home for the connected buildings ecosystem
Sign up for The Connected Buildings Briefing, our 5-minute briefing on the people, tools, and playbooks changing how building run. Delivered every two weeks to 8,000+ building professionals.
