Responsibilities
- End-to-End Project Ownership: Take comprehensive responsibility for development of computer vision pipelines, from initial concept to deployment including ideating, building and refining the solutions.
- High-Performance Pipelines: Spearhead the development of robust and scalable computer vision pipelines focusing on both local and cloud-based applications.
- Benchmarking and Working with State-of-the-Art Models: Engage in thorough benchmarking activities to assess and enhance the accuracy of latest computer vision breakthroughs. Strive for optimization and latest advancements in the field.
- Bug Fixing: Proactively identify and resolve software bugs, ensuring the deployment process is smooth and efficient.
- Cross-Team Collaboration for Solution Development: Engage in close collaboration with various stakeholders, including subject matter experts, to develop comprehensive solutions.
Technical Abilities
- Bachelor's or master's in computer science, Electrical Engineering, or related field, or equivalent experience.
- Over 2 years of experience working with Computer Vision, especially in training models and developing efficient pipelines, with a focus on object detection, image segmentation, and pose estimation.
- Proficiency with Python, image processing libraries like OpenCV and Pillow and deep learning frameworks like PyTorch and TensorFlow.
- Demonstrated expertise in developing and implementing computer vision pipelines, understanding visual data processing and algorithms.
- Solid understanding of data structures, algorithms, and software design principles.
- Experience in model optimization, including edge hardware deployment. Experience with Deepstream/GStreamer/Edge optimised libraries is a plus.
- Research experience or working experience with loT/Edge devices .
- Demonstrated ability to produce highly scalable pipelines, working effectively both independently and as part of a team.
- Excellent communication abilities to articulate findings and technical challenges to both technical and non-technical stakeholders.
