Currently, technically leading multiple computer vision teams covering projects in areas of document verification, traffic violation, crowd/individual behaviour monitoring, and healthcare. Personally engaged in the development of a smart document verification system using advanced deep-learning principles involving CNN, time-series 3D-CNN, and RNN/LSTM via the PyTorch platform.
Industrial experience: Android (+Native) .Net/C# (Microsoft Certified Specialist) C/C++ (CMake) Python (NumPy, SciPy, scikit learn, Pillow)
Cloud: Amazon Stack - AWS, EC2, API Gateways, Lambda functions, CLI/SDK with .Net/Python, Java (Android)
Computer Vision/AI/DL Frameworks/Methodologies: OpenCV, AWS Rekognition, Keras, PyTorch, TensorFlow, Kalman filters, CNN/LSTM/RNN systems and SVM
Web development Flask, Visual Studio/Azure SDK/IIS, Apache, LAMP
Continuous Integration/containerisation Docker, Anaconda
Robotics ROS, LCM
Previously led a team of data scientists, engineers and automotive specialists developing the next generation AI-driven autonomous systems. Experienced in end-to-end connected automotive vehicle system development.
Director of a consultancy firm specialising in the development of machine learning solutions for OCR and handwriting solutions for tax forms and other government documents.
Strong background in statistical domains including regression, clustering, random forests, and decision trees
A cumulative impact factor of 3.3004 in the domains of robotics, autonomous systems, IoT sensors and embedded electronics platforms.