Computer Vision Engineer

The Role:

As a Senior Computer Vision Engineer, you will own the entire computer vision pipeline, from dataset design to model deployment. You’ll be responsible for pushing the limits of detection, counting, tracking, and segmentation on large, complex, real-world road data.

Key Responsibilities:

• Dataset Management: Curate, label, augment, and manage large dash-cam datasets using RoboFlow for active learning and versioning.
• Model Research & Development: Build and fine-tune state-of-the-art CNN and Transformer models (e.g., YOLOv8/v9, SAM, RT-DETR) for object and scene understanding.
• Production Deployment: Convert and optimize models (using ONNX/TensorRT) for GPU servers and embedded edge devices, and set up CI/CD pipelines with Docker and Kubernetes.
• Performance Optimization: Define evaluation metrics, conduct A/B tests, and ensure high performance with a focus on achieving sub-100 ms latency and a mAP greater than 0.85 in real-world scenarios.
• Cross-Functional Collaboration: Partner with backend and frontend teams to integrate models as REST/gRPC APIs and create visual dashboards.
• Mentorship: Guide junior computer vision engineers and interns on best practices for data and model engineering.

Required Skills & Experience

• 3-6 years of hands-on experience in computer vision, with a proven track record of deploying models into production.
• Deep expertise with RoboFlow, including annotation workflows, versioning, auto-augmentation, and active learning.
• Strong proficiency in Python, PyTorch or TensorFlow, and OpenCV.
• Proven experience with model optimization techniques such as quantization, pruning, mixed-precision training, and TensorRT.
• Solid DevOps skills, including Git, Docker, CI/CD, and experience with cloud GPUs (GCP/AWS/Azure) or on-premise setups.
• Demonstrated track record of shipping CV products, published papers, or notable GitHub repositories.
• Experience with Edge AI platforms like NVIDIA Orin.
• Knowledge of real-time multi-camera synchronization and SLAM (Simultaneous Localization and Mapping).
• Experience with geospatial data, map matching, or traffic analytics.
• Familiarity with C++/CUDA for developing custom kernels.

Apply Now!
CTC up to 10 LPA
Experience: 5 years