Data Science Lead

About the Role
In this position, you will lead the design and development of cutting-edge computer-vision solutions encompassing object detection, tracking, re-identification, action recognition, and segmentation. You will be responsible for driving model selection and experimentation with advanced architectures such as YOLO, ViT/Transformer, and RF-DETR. Furthermore, your expertise in Vision-Language Models (VLMs) will be pivotal as you apply and guide their use, including fine-tuning practices. Your role will also involve optimizing models for production environments, ensuring they meet edge-AI constraints concerning latency, FPS, and GPU usage. As the owner of the full MLOps lifecycle, you will lead, mentor, and code-review a team, while contributing to solution architecture and effective communication with stakeholders.

Responsibilities

• Lead design and development of computer-vision solutions (object detection, tracking, re-identification, action recognition, segmentation)
• Drive model selection and experimentation with YOLO, ViT/Transformer architectures, and RF-DETR
• Apply and guide use of Vision-Language Models (VLMs), including fine-tuning
• Own model optimisation for production (TensorRT, ONNX, OpenVINO) within edge-AI constraints (latency, FPS, GPU)
• Own the full MLOps lifecycle end to end; lead, mentor and code-review a team; contribute to solution architecture and stakeholder communication

Qualifications

• Bachelor's or Master's degree (PhD a plus) in Computer Science, Machine Learning, Data Science or related field
• 10-15 years of applied machine-learning experience with deep computer-vision expertise
• Strong Python, with hands-on experience in PyTorch/TensorFlow
• Hands-on experience with YOLO, ViT/Transformer architectures, RF-DETR and VLM fine-tuning
• Demonstrated team leadership, mentoring, and stakeholder-communication / solution-architecture skills