Gaize had the science to detect drug impairment in real time but not the working technology — a prior contractor engagement left them without a functioning detection pipeline. We ran an extensive data-science program to extract reliable impairment signals from their high-frequency eye-tracking data,
Results at a glance
Detection Accuracy
Gaize set out to solve a problem that traditional drug testing can't touch: detecting whether someone is impaired right now, not whether they used a substance days or weeks ago. The science was sound — automated versions of the Drug Recognition Expert (DRE) ocular tests that law enforcement has relied on for over 45 years — but the technology to turn raw eye-tracking signals into trustworthy, defensible results didn't yet exist.
Before bringing us in, Gaize had invested in outside contractors to build that capability. The engagement left them empty-handed: no working detection pipeline, no reliable signal extraction, and no clear path from their VR-captured eye data to an impairment result a safety director — or a court — could stand behind.
The core difficulty wasn't a lack of data. It was the opposite. Each test produces an enormous volume of high-frequency ocular measurements, and the signals that distinguish impairment from normal variation are subtle, noisy, and easy to miss. Gaize needed a team that could find those signals and engineer them into a production-grade system.
We started where the previous effort fell short: in the data. We ran an extensive data-analysis and data-science program across Gaize's eye-tracking dataset — one of the largest of its kind, drawn from the world's largest cannabis impairment clinical trial and spanning hundreds of millions of data points — to isolate the ocular markers that reliably indicate impairment.
From there, we rebuilt Gaize's platform from the ground up:
The result was a system that doesn't just process data, but turns it into evidence safety leaders can act on with confidence.
FLIR Systems had run out of room to grow — literally. Their infrared sensor manufacturing and calibration lines were constrained by factory floor space and by the throughput and quality limits of manual labor. We led the project to fully automate their infrared camera assembly and calibration. The r
Book a 30-minute assessment—we'll map the bottleneck and the fastest path past it. No slide decks, no obligation.
Book your assessmentOur work took Gaize from a stalled prototype to a market-defining product. We enabled high-quality detection of both alcohol and cannabis impairment, and in doing so helped Gaize become the world's first commercially viable drug impairment detection technology — now trusted by industrial leaders across construction, mining, manufacturing, and logistics.
"DigiDaaS was able to deliver in 4 weeks what took our previous consultants 6 months"
VISIE's engineering team was swamped — their engineering and quality-control processes were almost entirely manual, capping how much the team could ship. We analyzed their ecosystem and requirements and delivered a thorough, mathematically derived and validated proposal and design to streamline both
