THE TEAM

Nitin Mane.

AI Architect/Engineer (Contractor)

Nitin is an artificial intelligence architect and engineer with over 4 years' experience designing, training, and deploying computer vision and edge AI systems for industrial and commercial operations.

He specialises in real time perception, taking models from dataset preparation and architecture through to inference running on the factory floor, with a particular focus on systems that hold accuracy under production conditions rather than benchmark ones. His work covers YOLO based detection and tracking, video analytics across continuous streams, and optimisation for constrained hardware through OpenVINO and NVIDIA Jetson.

He designs multi-modal systems that fuse vision, sensor telemetry, and machine logs into a single decision layer, using large language models and retrieval augmented generation to turn raw signals into alerts an operator can act on. His robotics work spans ROS2, embedded C++, and MATLAB and Simulink, including closed loop vision control of six degree of freedom manipulators.

He operates what he builds, working across Docker, Linux, CI/CD pipelines, and AWS infrastructure to take models from a research environment into supported production deployment. He is an M.Tech Mechatronics postgraduate of IIT Bhilai, an Intel Software Innovator and Intel Certified oneAPI Instructor, a Google Student Ambassador, an Arm Developer Ambassador, and Chair of the IEEE CEDA Nagpur Chapter.

History

Nitin has delivered applied artificial intelligence in industrial settings, including a real time anomaly detection pipeline for bar and rod mill video analytics, built on YOLOv8 and optimised through OpenVINO to run at line speed on plant hardware. His approach is grounded in validating detection performance against live operational footage rather than curated datasets, and in designing for the conditions that break vision systems in the field, poor lighting, occlusion, vibration, and hardware constraint.

His research is published and peer reviewed. Autograder+, a multi-faceted AI framework combining static code analysis with large language model driven pedagogical feedback, was published at ACM CODS 2025, and his earlier work on adaptive Kalman filter tracking with YOLOv5 was published in 2022. He contributes extensively to open source, maintaining a substantial public body of computer vision, robotics, and applied machine learning work.

He has held senior software engineering positions across a four year commercial career, and worked as an AI research and development engineer on medical imaging within the Peter Moss Leukemia AI Research Association, applying deep learning to haematology datasets. Earlier work as a MATLAB developer and research assistant established the signal processing and control foundations his robotics work is built on.

Nitin holds recognised expert status across the industry as an Intel Software Innovator and Intel Certified oneAPI Instructor, as a Google Student Ambassador, and as an Arm Developer Ambassador, representing those programmes internationally as a speaker and mentor. He serves the engineering community as Chair of the IEEE CEDA Nagpur Chapter and as a volunteer across IEEE SIGHT and IEEE Future Networks. He mentors developers, publishes openly, and works across the full path from research prototype to deployed system.

The Team

CREATE THE FUTURE

Let us build your future.

Bring us a plan or just an idea. We will arrange a discovery meeting with you to go through it.