LightMetrics – Deep tech, across the stack

The overarching vision of LightMetrics has been to make the benefits of AI-driven driver safety systems more widespread, starting from real-time driver coaching, all the way to event videos and advanced driver and fleet safety analytics available on the cloud for review by fleet managers.

Hiring for multiple roles

The overarching vision of LightMetrics has been to make the benefits of AI-driven driver safety systems more widespread, starting from real-time driver coaching, all the way to event videos and advanced driver and fleet safety analytics available on the cloud for review by fleet managers. A single set of RESTAPIs, supporting our Android, iOS and Linux SDKs that encapsulate AI inference on the edge, forms the backbone of our technology stack.

The breadth and depth of our technology stack is a direct consequence of this vision. In an era where every startup is deep tech, we believe in letting the scope of our products and services speak for themselves. Our products and services are driven by innovations across multiple domains, and we are proud to have some of the best engineering talents across multiple disciplines working in tandem to create amazing experiences for end-users.

AI and Computer Vision

Our AI team has an unenviable role – create a robust set of AI algorithms on the edge, for a variety of Advanced Driver Assistance System (ADAS) and Driver Monitoring System (DMS) use cases, while also being as compute and memory-efficient as possible, thus allowing our SDKs to be deployable across a wide variety of hardware platforms. Key challenges they grapple with and solve every day include:

  • Creating efficient neural networks that can process 10-15 fps video in real-time on compute and memory-constrained devices, by incorporating the latest in network quantization, pruning, etc.
  • Learning with minimal data, learning with noisy labels, ensuring minimal bias in learning are all hugely challenging, working on them and making things better, hugely satisfying
  • Highly optimized inference routines, using both host processors (NEON instruction set, fixed-point arithmetic, etc.), and  GPUs and AI accelerators
  • Handle variations in input video data across a multitude of parameters – time of day, weather, geography (different traffic signage), different types of vehicle and camera mounting heights, to name just a few

An ideal member of our AI team is not limited to training models on TensorFlow or PyTorch and moving on. They complete the entire journey from optimized model generation to implementation of inference routines on the device, and finally to large scale deployment across our global installed base of vehicles.

Web services, APIs and Front-end

Our customers live and die by our web services and APIs, so their scalability and uptime are of utmost importance to us. Every member of our backend team is a full stack developer, working across the entire breadth of our platform:

  • Designing and extending our microservices that support our APIs, to scale with the growth within our existing customer base, and also new customers that we keep adding at a rapid clip
  • Creating and managing RESTAPis that export the data and analytics that our customers need to create the best experience for their end-users on their dashboards
  • Harnessing the data we are gathering at scale to create new services that enhance our solutions, and create paradigm-shifting end-user experiences
  • Working with our AI and SDK teams, to ensure that our APIs grow in lock-step with new features they develop
  • Creating front-end visualizations and interactions that help demonstrate the ability of our APIs to power amazing user experiences, that our customers can learn from and incorporate in their own dashboards
  • Incorporating changes to our backend as we scale across geographies and need to support privacy regulations (GDPR), and other compliance requirements
  • Creating our API roadmap with feedback from our community of developers

SDK (Android, iOS, and Linux)

Our SDK is the beating heart of our product. It is where all the different parts of our technology stack come together – interfacing with our AI libraries, communication with our backend web services, and providing APIs for application developers to access real-time alerts and other data they can use to create engaging and effective mobile user experiences.

Success in this role is predicated on a deep knowledge of Android/iOS/Linux development, systems engineering, and the ability to write production-quality code in multi-threaded environments. It also requires a good understanding of AI inference workflows and working closely with our AI team to ensure the efficient integration of new models and features.

This is also a customer-facing role – for a growing community of external developers who rely on our SDK to build applications and services, this team is responsible for guidance and integration support. In addition, a significant part of our development roadmap comes from interactions with our community of developers.


We have a very active internship program and offer 3-month and 6-month internships for students enrolled in undergraduate and postgraduate programs in relevant engineering disciplines. Interns never work on ‘dummy’ projects, and their work is always mapped to things in our roadmap that our product teams are not yet ready to take on. In fact, many features in our current roadmap have evolved from past intern projects. Past interns have worked on all aspects of our technology stack – AI, client SDK, and web services.

An internship at LightMetrics is a great way for students to bolster their resumes and find their calling – a natural progression to joining our engineering team (which we like a lot), grad studies or jobs at leading corporations. Either way, we are glad to be the launching pad for amazing new careers!

LightMetrics is growing, and we need talent across the entire spectrum of our stack. If a fast-paced development environment that gives you the opportunity to see your work deployed in the field and making a meaningful difference motivates you, reach out to us at