Netradyne
About this role
Netradyne puts computer vision on the edge in commercial vehicles, analysing driving in real time to flag unsafe behaviour before a collision happens, and says it has vision analysed more than 25 billion miles. This role is on the Analytics team and the work is genuinely applied research: designing, implementing and commercialising driver monitoring and driver assistance algorithms, with access to very large datasets and deployment into thousands of vehicles. That deployment path is the interesting part, because the models have to run on constrained edge hardware rather than in a data centre. At 5 or more years this is the more accessible of two near identical Netradyne research postings, the other being a Senior Staff version asking 9 or more years.
Who this is for
The employer asks for 5 or more years of research or industry experience within computer vision and machine learning models. Note that Netradyne writes this as "5+ yrs", abbreviated, which is worth mentioning because automated parsers frequently miss that form.
Required: a BTech, MTech or PhD in computer science, electrical engineering or a related area; in depth understanding of machine learning covering both deep learning and classical ML, and of computer vision concepts; excellent programming skills in Python, which is required, with C++ desired; the ability to turn abstract product concepts into working systems; the ability to innovate on problems without documented solutions; and hands on experience with edge deployments targeting various hardware platforms.
Preferred: experience with time series data, camera calibration, multi view geometry, 3D reconstruction and SLAM, and road scene understanding covering objects and lanes.
The day to day: designing and commercialising algorithms that characterise driving behaviour, deploying new algorithms and models into thousands of vehicles, and collaborating cross functionally to build and deploy scalable AI systems.
Location is Bangalore. Netradyne is consistently the one employer on this board that publishes its office policy, and its other current postings state onsite five days a week at Whitefield on a general shift.
Good fit if you have computer vision depth and want models that ship to real hardware. Not a fit if you want purely offline research, or cannot work onsite in Bangalore.
Required: a BTech, MTech or PhD in computer science, electrical engineering or a related area; in depth understanding of machine learning covering both deep learning and classical ML, and of computer vision concepts; excellent programming skills in Python, which is required, with C++ desired; the ability to turn abstract product concepts into working systems; the ability to innovate on problems without documented solutions; and hands on experience with edge deployments targeting various hardware platforms.
Preferred: experience with time series data, camera calibration, multi view geometry, 3D reconstruction and SLAM, and road scene understanding covering objects and lanes.
The day to day: designing and commercialising algorithms that characterise driving behaviour, deploying new algorithms and models into thousands of vehicles, and collaborating cross functionally to build and deploy scalable AI systems.
Location is Bangalore. Netradyne is consistently the one employer on this board that publishes its office policy, and its other current postings state onsite five days a week at Whitefield on a general shift.
Good fit if you have computer vision depth and want models that ship to real hardware. Not a fit if you want purely offline research, or cannot work onsite in Bangalore.
Apply on company site
Opens www.netradyne.com, the employer's own application page. Applying is always free.