Computer Vision Engineer

Key Responsibilities

• Build end-to-end computer vision solutions from data preparation and model development to deployment and monitoring.
• Develop and fine-tune object detection, segmentation, classification, and tracking models.
• Work with PyTorch for computer vision and deep learning model development.
• Evaluate models using mAP, IoU, Dice, Precision, Recall, and F1 Score.
• Develop real-time video processing pipelines using OpenCV, GStreamer, and FFmpeg.
• Optimize models for real-time inference using ONNX, TensorRT, and OpenVINO.
• Develop and integrate REST APIs using FastAPI or Flask.
• Deploy applications and ML/CV pipelines in Linux environments.
• Use Docker for containerized deployments where applicable.
• Debug and troubleshoot models and video processing pipelines using metrics, logs, and performance analysis.
• Collaborate with backend and engineering teams to integrate CV components into production systems.
• Contribute to monitoring and logging of deployed ML/CV systems.

Mandatory Skills

• 2+ years of hands-on experience in Computer Vision / Machine Learning Engineering.
• Strong programming skills in Python.
• Strong experience with PyTorch.
• Hands-on experience with at least one of:
• Object Detection
• Image Segmentation
• Object Tracking
• Image Classification
• Good understanding of mAP, IoU, Dice, Precision, Recall, and F1 Score.
• Strong experience with FastAPI for API development.
• Experience deploying applications/models on Linux.
• Experience with OpenCV.
• Strong model and pipeline debugging skills.
• Good problem-solving and communication skills.
• Ability to work in a research-driven and hands-on engineering environment.

Good to Have

• GStreamer, FFmpeg, and RTSP streaming.
• ONNX, TensorRT, and OpenVINO.
• Edge deployment experience with NVIDIA Jetson, ARM, or Embedded Linux.
• Experience with YOLO, Detectron2, ViT, Swin, SAM, or CLIP.
• Docker and CI/CD experience.
• Experience with CVAT or Label Studio.
• Familiarity with MLflow or Weights & Biases.
• Experience with ML/CV monitoring and logging.
• Exposure to LLMs, Speech AI, or edge AI systems.

Preferred Candidate Profile

• Strong hands-on experience building production-grade Computer Vision solutions.
• Comfortable working across model development, APIs, video pipelines, and deployment.
• Ability to optimize CV models for real-time inference and edge environments.
• Strong debugging and problem-solving approach.
• Comfortable working in a research-oriented environment while delivering production solutions.