Bengaluru Engineer Builds App That Spots Potholes And Identifies Who Must Fix Them
Summarized by AI; it may make mistakes. Check important info
Summarized by AI; it may make mistakes. Check important info
For most Bengaluru motorists, a pothole means a rough ride, possible vehicle damage and another complaint to the civic authorities. Engineer Gaurav Sen has taken a different approach, building an app that can not only detect potholes but also identify the contractor responsible for repairing the affected stretch of road.
Sen demonstrated the system in a video shared on Instagram in collaboration with ChatGPT India. The video later circulated on X, drawing attention to his attempt to turn a common road problem into a trackable civic issue.
How the pothole detection system works
Sen fitted his car with a dashcam, GPS and an accelerometer to collect road data while driving. The footage is then analysed using computer vision to identify potholes and assess their size.
According to Sen, he used Codex to build the application, which records the relevant information and sends it to OpenAI for processing. The system is also designed to distinguish potholes from features such as speed breakers, which can sometimes appear similar from a distance.
But spotting the damage was only part of the problem Sen wanted to solve.
App searches 2,900 government contracts
The engineer modified the system to determine which contractor is responsible for a particular road.
The app searches through around 2,900 government contracts to match a detected pothole with the relevant road contract. This could be particularly useful for roads still covered by a warranty, where contractors may be required to repair defects without the government having to pay additional costs.
The system can reportedly generate a structured record in about four seconds. The record includes a photograph of the pothole, its precise location, the relevant tender number and the officer responsible for the road.
System detected 12 potholes on one commute
Sen also tested the system during a drive to work. It identified 12 potholes along his route, providing documented information that could potentially be used while filing complaints with civic authorities.
The project has since attracted attention beyond Bengaluru, with users on X praising the idea and suggesting that similar technology could help civic bodies identify damaged roads and prioritise repairs.
What started as an attempt to make one engineer’s daily commute less frustrating has developed into a practical experiment in making pothole complaints more specific, documented and easier to route to the authority or contractor responsible.
