Position - Machine Vision/Computer Vision Engineer
Location - Chennai
Experience - 2+ years
Key Responsibilities
- Algorithm Development: Design, implement, and optimize computer vision and deep learning algorithms for object detection, segmentation, 3D reconstruction, OCR, anomaly detection, and pose estimation.
- End-to-End Pipeline Ownership: Develop and maintain robust vision pipelines, encompassing data ingestion, preprocessing, inference, post-processing, and system integration.
- Model Lifecycle Management: Train, fine-tune, evaluate, and deploy deep learning models (CNNs, Transformers, etc.) using frameworks like PyTorch, TensorFlow, or JAX.
- Data Strategy: Lead dataset curation, annotation, and augmentation efforts, including the use of synthetic data generation to improve model robustness.
- Testing & Validation: Conduct rigorous testing, benchmarking, and error analysis under real-world conditions to ensure system reliability and performance.
- Performance Optimization: Optimize models for edge deployment using techniques such as quantization, pruning, and leveraging inference engines like TensorRT, ONNX, or OpenVINO to meet strict latency and power constraints.
- Embedded Integration: Collaborate with embedded engineers to port and optimize vision software for various hardware platforms (SoCs, GPUs, NPUs).
- Research & Innovation: Stay current with the latest research in computer vision and machine learning, rapidly prototyping new techniques to enhance our capabilities.
- Software Engineering Best Practices: Write clean, documented, and maintainable code, and actively contribute to internal libraries, tools, and code reviews.
Required Qualifications
- Bachelor's/master's in computer science, Electrical Engineering, Robotics, or a related field.
- 2+ years of hands-on industry experience in computer vision or machine learning (including relevant internships).
- Strong programming proficiency in Python and C++.
- Deep, practical experience with at least one major deep learning framework (PyTorch is strongly preferred).
- A proven track record of deploying real-time computer vision systems into a production environment.
- Solid foundation in classic computer vision (OpenCV, camera calibration, feature detection, multi-view geometry).
- Experience working with 2D/3D data formats, point clouds, and sensor fusion (e.g., camera with LiDAR/IMU).
