Bengaluru engineer uses AI to turn pothole detection into ready-to-file complaints

A Bengaluru-based engineer has developed something as simple as AI support to fix potholes that can help citizens raise civic complaints and demand accountability from both government officials and contractors responsible for the work.

In Short

• Dashcam, GPS and accelerometer data help the system record road conditions
• AI vision classifies potholes by size and filters out speed breakers
• The software searches about 2,900 contracts to trace responsible agencies

A Bengaluru-based software engineer and entrepreneur has shown how artificial intelligence can be used to solve a problem that millions of Indians face every day, potholes.

Gaurav Sen, founder of InterviewReady, recently demonstrated an AI-powered system that can detect potholes while a car is being driven, identify their location, analyse their size and then generate a ready-to-file civic complaint.

The project has attracted attention online because it goes beyond simply finding a pothole and attempts to connect the problem with the government contract and contractor responsible for the road.

HOW THE AI POTHOLE SYSTEM WORKS

The system uses a dashcam along with GPS and an accelerometer to record the road while the vehicle is moving. AI vision is then used to identify potholes and classify them according to their size, while attempting to distinguish them from features such as speed breakers.

The more interesting part comes after the pothole is detected.

According to the demonstration shared online, the system can search through roughly 2,900 government contracts to identify the tender connected to the road, determine the contractor responsible and find the relevant government officer.

It can then attach the pothole's photograph and geographical coordinates and prepare a structured complaint.

In the example circulated online, one drive detected 12 potholes, with the system producing 12 complaints that were ready to be filed.

The idea is particularly relevant to Bengaluru because the city's civic authorities already operate systems for reporting potholes and other public infrastructure problems.

The former BBMP's official "Fix My Street" system, for example, allowed citizens to submit geo-tagged photographs of potholes and send complaints to the concerned ward officials.

The current Greater Bengaluru Authority website also lists a "Fix Pothole" service among its civic applications.

Government documents also show that road and pothole repair work is regularly assigned through contracts, with provisions relating to maintenance and defect-liability periods.

WHY THE PROJECT IS GETTING ATTENTION

The demonstration has been widely discussed across social media, with users especially the Gen Z questioning how far AI can be taken beyond chatbots and content generation.

We recently witnessed how the two young generations are questioning the existing system and demanding a change that could reflect in the result.

Discussions around the project have also raised a practical question: if AI can identify a pothole, locate it and connect it with government records, could similar systems be built for other everyday civic problems?

The response online has been mixed, with some users calling for the idea to be expanded to other Indian cities.

Indian cities are widely criticised for their infrastructure problems, with netizens raising questions in large numbers and demanding at least accountability and action. The bureaucracy is at the centre of these concerns, while politicians are also being questioned for delivering on the responsibilities for which they were elected.

Sen shared the development through his online presence, including LinkedIn, where his profile identifies him as the founder of InterviewReady and records his previous experience as a software engineer at Uber, a platform engineer at Directi and an IT analyst at Morgan Stanley.

His LinkedIn profile also lists an accelerator programme in entrepreneurship at Cornell University and his bachelor's degree in computer science from Fr Conceicao Rodrigues College of Engineering.

The pothole project therefore offers a more practical example of what AI can look like outside the usual demonstrations. A camera sees the road, AI understands what it sees, government data provides context, and software turns the result into an actionable complaint.

For a city where potholes remain a familiar part of daily travel, that is a simple idea with potentially much bigger implications.

- Ends