Loading...
Interview with Anup Patil, CEO and Co-Founder of Intangles
India’s commercial vehicle market is entering a new phase of growth, supported by infrastructure investment, fleet renewal, electrification and stronger safety requirements. At the same time, operators are looking beyond basic vehicle tracking towards systems that can connect uptime, fuel efficiency, safety and driver performance.
In this interview, Anup Patil, CEO and Co-Founder of Intangles, discusses the role of AI-powered fleet intelligence, Digital Twins, ADAS and cross-OEM data in the future of commercial vehicle operations.
The Indian commercial vehicle segment is experiencing broad-based growth. Infrastructure investment, State Transport Undertaking replacement cycles, route expansion and electric-bus programmes are all supporting demand.
At the same time, fleet operators are changing the way they evaluate technology. They are no longer asking whether digital tools are necessary. They want to know which technologies can connect fuel consumption, vehicle uptime, safety and driver behaviour in one system.
Safety regulations are also influencing procurement and engineering decisions. Requirements covering emergency braking, lane-departure warnings, drowsiness detection and blind-spot monitoring are increasing the importance of integrated vehicle intelligence.
Intangles has operated in this market since 2016. The company says its platform is currently used across 18 countries and more than 500,000 vehicles, with over 40 OEM and enterprise partners.
In India, its partnerships include Mahindra Truck and Bus, Force Motors and electric-bus manufacturers. The platform is used across freight, logistics, state and intercity bus operations, school transport, construction, mining, agriculture, cold chain, last-mile delivery and municipal fleets.
A single temperature, voltage or fuel reading has limited value without context. The important question is whether that reading is normal for a particular vehicle under its current operating conditions.
Each vehicle connected to our platform has its own Digital Twin. This virtual model is continuously updated using data from onboard sensors and the electronic control unit. It considers mileage, operating history, driver behaviour, wear and duty cycle.
This allows the system to establish an expected operating pattern for each vehicle and identify deviations before a fault becomes visible through the vehicle’s standard systems.
For internal-combustion fleets, the same approach is used for fuel monitoring. Raw fuel-level readings can be distorted by tank geometry, gradient and vehicle movement. Machine-learning models help distinguish genuine fuel movement from normal sensor fluctuations.
For electric fleets, the Digital Twin supports range prediction by analysing motor torque, wheel speed, weather, HVAC and lighting loads, driver behaviour and battery degradation history. It also monitors battery-coolant temperature, cell-level voltage and charging patterns.
According to Intangles, its predictive models operate at 96% AI accuracy, helping operators move from monitoring past events to anticipating future issues.
A cross-OEM platform can identify patterns that may not be visible within a single manufacturer’s ecosystem.
Bus operators often manage vehicles from several OEMs across state routes, school transport, intercity services and electric-bus operations. When data remains separated by manufacturer, operators must manually reconcile different systems and performance indicators.
Data sharing can benefit all parties. OEMs gain access to real-world performance information from deployed vehicles, which can support product development and warranty planning. Operators receive a more complete view of fleet performance, while technology providers can analyse patterns across different manufacturers and operating conditions.
However, this requires clearer frameworks for consent, data ownership, privacy and access. Until those areas are more fully defined, the value of industry-wide data sharing will remain partially unrealised.
The common problem is that critical information often arrives too late.
Unplanned downtime is one of the clearest examples. A breakdown can create repair, towing, replacement-vehicle and service-delay costs. Predictive systems can identify potential problems earlier, allowing operators to schedule maintenance rather than react to a roadside failure.
Fuel loss presents a similar challenge. Siphoning, underfilling, adulteration and unauthorised refuelling may remain undetected until they appear in a monthly report. Real-time fuel analysis can give operators the location, time and estimated volume of a suspicious event.
Driver behaviour is another major factor. Harsh braking, inefficient acceleration, idling and overspeeding affect fuel consumption, maintenance requirements and accident risk. Driver-profiling systems can connect these behaviours with their operational consequences, allowing more targeted coaching.
A growing challenge is managing mixed electric and diesel fleets. Electric vehicles have different health indicators, range variables and failure risks. Operators need tools that can manage both powertrains while maintaining the correct technical indicators for each.
The goal is to make commercial vehicles more transparent to operators, OEMs and service networks.
The platform already correlates signals across several vehicle systems to identify anomalies and recommend possible interventions. The next stage is to expand this intelligence from the individual vehicle to the wider fleet network.
At network level, the system could identify a component-stress pattern across a specific engine configuration, a route condition that accelerates wear, or a charging practice that gradually reduces battery life.
The long-term objective is to automate more of the maintenance process. A platform could identify an issue, recommend the required intervention, schedule it around the operator’s route plan and coordinate with the service network to ensure that the correct parts and technicians are available.
For OEMs, data from deployed vehicles could provide detailed insight into performance across different geographies, duty cycles, driver profiles and environmental conditions.
Intangles aims to position itself as the intelligence infrastructure connecting vehicles, operators, OEMs and service networks. The company raised USD 30 million in Series B funding in October 2025, led by Avataar Venture Partners, with continued backing from Baring Private Equity India and Cactus Partners.
ADAS can directly address several major risks involving heavy buses and trucks, including fatigue, delayed driver reaction and blind-zone incidents.
Automatic emergency braking is especially important for heavy vehicles because of their longer stopping distances. It can intervene when a driver does not react quickly enough to an obstacle.
Drowsiness detection is particularly relevant for long-distance and night operations. Driver fatigue can be influenced by route length, working hours, heat and repeated shifts. Systems that monitor eye behaviour and steering patterns can identify signs of impairment before they lead to an accident.
Blind-spot monitoring is important in urban environments where pedestrians, cyclists and two-wheelers may enter areas that the driver cannot see during low-speed manoeuvres.
Intangles’ Video Telematics platform processes alerts inside the vehicle rather than relying entirely on a cloud connection. It detects distraction, drowsiness, lane drift and unsafe following distance, and provides an in-cabin warning.
Events are recorded with location, time, driver identity and video footage. Intangles reports a 20–30% improvement in driver behaviour among fleets using Driver Profiling and Video Telematics together.
The most difficult fuel losses are often internal operational losses rather than changes in fuel prices.
These include siphoning, adulteration, unauthorised refuelling and discrepancies between recorded and delivered quantities. Because they happen gradually, they may remain unnoticed for weeks.
Fuel measurement is complicated by tank shape, vehicle gradient, movement and parking angle. Conventional threshold-based systems can generate false alerts or fail to detect genuine events.
Intangles uses machine-learning models to distinguish environmental noise from actual fuel movement. The system can identify the time, location and estimated volume of a fill or drain event.
The company says it uses existing OEM sensors, without requiring additional tank hardware or modifications to the fuel system.
Monitoring fuel cost per kilometre at vehicle level can also reveal mechanical inefficiency. A vehicle consuming more fuel than comparable units under similar conditions may be developing a technical issue before a fault code appears.
Fuel performance is also connected to driver behaviour. Harsh acceleration, excessive idling, coasting in neutral and overspeeding all increase consumption. Combining fuel analysis and driver profiling in one platform allows operators to address both mechanical and behavioural inefficiencies.
