AI in Automotive Podcast · Jayesh Jagasia

AI in Automotive - #404 - David Hallac - CEO, Viaduct

·41 min·1 clip

The episode opens on season four and the broader AI-and-mobility frame. The host starts with a familiar automotive problem, then asks why vehicle quality issues still cost OEMs so much when connected data is everywhere. Recalls, repairs, lawsuits, and lost revenue can add up to tens of billions of dollars for manufacturers each year. The guest is David Halak, CEO of Viaduct. The company finds patterns and relationships among billions of connected vehicle data and delivers two practical use cases to automotive OEMs. One is proactive quality detection, which can help OEMs address problems before they become large recall events. The other is failure prediction. That use case supports proactive maintenance and better uptime, especially for fleet customers. The conversation then moves into how the system works. The guest says the underlying algorithm is generic and can be applied across tire, transmission, battery, aftertreatment, and other issue types. The differences sit elsewhere. Most of the variation lives in the data model and the application layer, where the system is shaped to the component and the user workflow. He breaks the stack into three cloud layers plus the edge layer. The edge layer collects data from compute, sensors, and the vehicle platform itself. The data layer is opinionated. It is built around automotive-specific ways of deciding what to collect and how to collect it, with tens of thousands of automotive-specific features already mapped across components and systems. The application layer turns the findings into something users and experts can act on. The episode ends on the idea that the same relationships can surface across vans, trucks, light commercial vehicles, and cars, which is why the method feels foundational rather than brand-specific.

As heard by us

Connected-vehicle AI framed as quality control and fleet uptime.

The episode keeps its focus on one practical question: why connected-vehicle data matters when it can cut recall, repair, and lawsuit costs for OEMs.

Read the full review in PlayNext →

Why you'd press play

You want connected-vehicle AI that turns data into practical savings before it gets flashy.

Read the full recommendation in PlayNext →

Listen to the show on

Apple Podcasts Spotify

Play clips from this episode More from AI in Automotive Podcast