Electric vehicles are usually discussed in terms of range, charging speed and battery capacity. Yet a vehicle’s usefulness also depends on what happens after it leaves the showroom. Understanding how its battery behaves through repeated charging cycles, changing temperatures and daily use is an important part of keeping it operational.
Battery monitoring provides information about present conditions. The wider opportunity is to use that information over time, identifying patterns that help engineers investigate problems and make better maintenance decisions.
This is the engineering challenge behind Siba Shankar Panda’s interest in EV battery intelligence. A Senior Full-Stack Engineer and Project Lead with more than a decade of software experience, Panda works across backend services, cloud infrastructure and user-facing applications. His experience includes EV telemetry systems and an association with battery-software startup Nicola Charging.
What makes his work worth examining is the connection between those different layers. Discussions of artificial intelligence often centre on the model itself. Panda’s experience brings attention to the surrounding software that receives data, processes it and presents the results to people who need to act.
During his employment with Tata Consultancy Services, Panda worked on a Tata Motors engagement involving EV telemetry and battery-health monitoring. According to his account of the project, his contribution included an MQTT-based backend processing more than 50 telemetry packets per second from a fleet of thousands of EVs.
MQTT is a messaging protocol used to exchange information between connected devices and software systems. In this setting, it helped connect vehicle-generated readings with the services responsible for processing and displaying them.
Panda’s project account also describes work with TensorFlow Lite models for battery anomaly detection and predictive-maintenance use cases. Alongside this, he developed React-based dashboards for telemetry, alerts and device monitoring.
Together, these components created a route from vehicle data to information an operational team could investigate. The backend received the readings, analytical models helped identify unusual patterns, and the interface made those outputs accessible to users.
This is where full-stack engineering becomes particularly relevant. A useful monitoring application needs dependable connections between its components. Engineers must consider how data arrives, how it is handled and how clearly the resulting information is presented. A technically sophisticated model has limited operational value if its findings never reach the right person in an understandable form.
Panda’s association with Nicola Charging provided another perspective on the same problem. The company’s product materials describe cloud-based battery enterprise software supporting remote monitoring, battery diagnostics, vehicle telematics and remote configuration of Battery Management System parameters.
His documented role at Nicola focused on frontend development, including analytics interfaces and integration between dashboards and backend services. His role also included an equity arrangement offering a 2% stake, subject to vesting, alongside these technical responsibilities.
These responsibilities placed him at the point where battery information became available to a software user. The equity arrangement added a potential ownership interest to his association with the company developing the platform.
The two settings connect around a common question: how can information from a physical battery become useful within a wider operational system? The Tata Motors engagement involved vehicle telemetry and monitoring workflows. Nicola’s product environment approached battery management through remotely accessible enterprise software.
For Panda, these experiences inform a broader view of batteries as assets with a continuing digital history. In the perspective he has shared, the value of telemetry extends beyond a reading at a particular moment. Its greater potential lies in helping engineers understand how behaviour changes over time.
An evolving battery-health profile could bring together charging history, temperature, usage patterns and anomaly signals. With sufficient data and validated models, it could help teams investigate whether degradation is accelerating, which operating conditions accompany unusual behaviour and when closer inspection may be appropriate.
Such capabilities would have to earn users’ confidence. An unusual reading does not automatically reveal a fault’s cause. Predictive systems need careful testing, reliable data and technical oversight. Their outputs must also be clear enough for maintenance teams to distinguish an observation from a recommendation.
Panda sees the longer-term opportunity in connecting these insights with fleet decisions. Battery-health information could be considered alongside charging schedules, maintenance records and vehicle utilisation. Over time, that could support more informed decisions about when a vehicle should be inspected or how an asset is managed throughout its working life.
Achieving that vision will require cooperation between software engineers and battery specialists. Cloud infrastructure, analytical models and interfaces each address part of the problem; their usefulness depends on how well they work together.
Panda’s full-stack background offers a practical perspective on this connection. His work links the systems receiving vehicle data with the applications through which people interpret it. As battery software develops, that ability to connect physical measurements with usable information will remain central to the move from monitoring towards predictive intelligence.
