At Motive, some of the most impactful engineering work happens behind the scenes. For Saravana Mahesh, that work starts with a deceptively simple question: Who was that driver behind the wheel?
As the engineering lead for backend systems on Motive’s Driver ID platform, Saravana helps power the infrastructure that identifies which driver was behind the wheel for every trip, safety event, and vehicle movement across Motive’s platform.
That driver identity is what allows fleets to automatically assign trips, accurately coach drivers after unsafe behavior, maintain reliable hours-of-service records, and improve accountability for vehicle usage and fuel spend.
Turning Face Match into a real-time platform capability
One of Saravana’s recent contributions was helping launch Real-Time Face Match*.
*Face Match technology automatically identifies drivers and matches trips to help teams save time, reduce unassigned trips, and improve hours of service (HOS) compliance. Not available in Illinois.
Previously, Face Match identified drivers post-trip. Saravana redesigned the backend so identification now happens in motion—giving fleets real-time driver attribution.
To enable this, the team rebuilt the pipeline end-to-end:
- Ingest face detection events directly from the vehicle
- Trigger immediate image recall
- Match drivers using AWS Rekognition
- Write identities to the Driver ID Service, creating a single, consistent source for downstream products
This transformed Face Match from a standalone feature into a shared platform capability used across Motive.
The shift introduced significant technical challenges, particularly reducing latency in a system built on long, fragile synchronous processes. By moving to a resilient, event-driven architecture with stronger observability, the team cut processing time from 16 minutes to 7—a 56% reduction.
This improvement gives fleets driver attribution fast enough to act—whether reviewing a safety event, tracking a trip, or analyzing fuel activity.
Solving customer problems at the platform level
Some of Saravana’s most meaningful work started with a single support ticket.
When a fleet in Mexico reported that drivers weren’t being identified, Saravana discovered the root cause: U.S.-specific privacy logic in the Face Match pipeline was breaking identification in other regions.
Rather than treating it as an isolated bug, he identified it as a systemic gap that was impacting every fleet operating in Mexico.
Saravana led the effort to redesign the system’s privacy logic to support country-aware restrictions, allowing Driver ID to adapt to regional requirements without re-engineering the platform.
The result: Face Match recall rate for fleets in Mexico jumped from 38% to 85%.
That work didn’t simply resolve one customer issue — it unlocked international scalability for Driver ID and improved Motive’s ability to deliver consistent safety and compliance workflows across regions.
Engineering for reliability at scale
Saravana’s team owns one of the most foundational systems in Motive’s platform, where reliability is critical.
A safety event can’t be coached unless the right driver is identified. Fuel accountability, compliance workflows, and emergency response all depend on accurate driver attribution.
Saravana recently led a redesign of the face detection pipeline behind real-time Face Match to make it faster and more reliable.
Before, image recall and identity assignment ran through a synchronous system that slowed down under load. By rebuilding it as an event-driven architecture with clear queues and backpressure handling, the team significantly improved performance and resiliency.
They also invested in:
- Tighter reliability targets and monitoring
- Automated cleanup for pipeline failures
- Stronger test coverage across identity write paths
- Local development environments that reduce debugging time from days to hours
These backend improvements may not be visible to customers, but they directly improve the consistency and trustworthiness of the products fleets rely on every day.
Building AI workflows for engineers
Saravana says one of the most exciting parts of working at Motive is how his team uses AI — not as a novelty, but as a critical part of the engineering workflow.
His team has built internal AI workflows that help:
- Break design docs into scoped engineering tickets
- Guide implementation with built-in checks
- Resolve pull request feedback
- Clean up legacy code after rollouts
This AI-first development approach helps engineers spend less time on repetitive work and more time on architectural decisions and customer impact.
That investment in engineering velocity is helping the team build faster while maintaining quality — especially important in a platform that operates at the intersection of edge AI, distributed systems, and global privacy requirements.
Why our work matters
For Saravana, the most rewarding part of engineering at Motive is seeing how backend infrastructure improvements translate directly into better outcomes for customers.
When Driver ID is accurate, fleets can automatically assign trips to the right drivers without manual cleanup. Safety teams can immediately connect risky driving events to the correct person, making coaching faster and more effective. Compliance teams can rely on accurate driver records for hours-of-service reporting, reducing administrative burden and lowering audit risk. And with greater visibility into who is operating each vehicle, fleet managers can improve accountability across safety, fuel usage, and day-to-day operations.
These improvements may happen behind the scenes, but the impact is immediate: safer fleets, less manual work, stronger compliance, and more confidence in the data teams rely on every day.
That’s what makes the work so meaningful. At Motive, engineering teams aren’t just building backend systems — they’re building the intelligence layer that helps the physical economy run safer, smarter, and more efficiently.
