Bengaluru Engineer Builds AI Tool to Track Potholes & Contractors (Image Courtesy: Instagram/@gkcs_)
NEW DELHI: A Bengaluru-based engineer has developed an AI-powered application that not only detects potholes on roads but also identifies the contractor responsible for their repair and prepares ready-to-file complaints for civic authorities.
The system, created by Gaurav Sen, has gained widespread attention after a demonstration video went viral on social media.
Sen, who is also the founder of InterviewReady, said frequent road accidents and vehicle damage caused by potholes motivated him to build the solution.
How the System Works
The app uses a combination of a dashcam, GPS and an accelerometer fitted in a vehicle to collect data while driving. The footage and sensor information are processed using AI vision models to identify potholes of different sizes, including those that may resemble speed breakers and are difficult to spot from a distance.
Sen used Codex to develop a small application that records the relevant values and sends them to OpenAI for analysis. The system can distinguish potholes from other road features and classify them based on size.
Linking Potholes to Contractors
Beyond detection, the app searches through approximately 2,900 government contracts to determine which contractor is responsible for a particular stretch of road.
This feature is especially useful when roads are still under warranty, as the contractor may be obligated to carry out repairs without additional cost to the authorities.
Within four seconds, the system generates a structured output that includes:
• Photograph of the pothole
• Exact geolocation
• Tender or contract number
• Name of the officer responsible for the road
This information can be used to prepare formal complaints for civic bodies.
During one of his drives to work, Sen’s system detected 12 potholes along the route. Each detection produced documented evidence that could be submitted to the relevant authorities.
The demonstration video, shared on Instagram in collaboration with ChatGPT India and later circulated widely on X, has drawn praise from users.
Many described the concept as “fantastic” and suggested that state governments could adopt similar technology for better awareness and faster response to road defects.
Others noted that while the technology is impressive, the real challenge remains getting authorities to act on the complaints.
The project has been viewed as an example of how artificial intelligence can be applied to everyday civic problems, turning individual frustration with poor road conditions into a data-driven tool for accountability.
About the Author
