Senior Data Scientist

We are always looking for amazing talent who can contribute to our growth and deliver results! Geotab is seeking a Senior Data Scientist who will design and build production-ready analytics systems that extract actionable insights from video, image, and sensor data. The projects will vary in scope, complexity, and affected business area. If you love technology, and are keen to join an industry leader — we would love to hear from you!

As a Senior Data Scientist, your key area of responsibility will be evaluating and integrating Vision Language Models (VLMs) into scalable cloud-based processing pipelines, developing robust evaluation frameworks to measure model quality across diverse product use cases, and optimizing end-to-end inference performance for cost, latency, and accuracy. You will need to work closely with Geotab’s product and platform teams to deliver computer vision features across safety, compliance, and geospatial applications.

To be successful in this role, you will be a strong team-player with an adaptive mindset to stay current with the rapidly evolving multi-modal AI landscape and the ability to engage effectively at all organizational levels. In addition, the successful candidate will have strong project management and analytical skills, with the ability to identify needs, develop effective solutions, and manage projects from research through production deployment.

• Lead the research, evaluation, and integration of Vision Language Models (VLMs) and foundation models to extract actionable insights from video, image, and telematics data.
• Design experimentation frameworks to systematically benchmark model performance across accuracy, precision, recall, and domain-specific quality metrics.
• Develop prompt engineering and fine-tuning strategies for multi-modal reasoning tasks, optimizing for visual understanding in real-world operational environments. Build scalable evaluation methodologies that replace manual labeling with automated, data-driven assessment approaches.
• Combine video, image, and telematics data to develop contextual models that enrich visual insights with sensor telemetry, vehicle dynamics, and geospatial signals.
• Collaborate with cross-functional product teams to translate business problems into well-scoped data science solutions, from hypothesis through production deployment.
• Partner with engineering teams to ensure models are production-ready, balancing inference cost, latency, and accuracy at scale.
• Stay current with advances in vision-language model research and drive adoption of emerging techniques into the product portfolio.
• Mentor team members on best practices for vision model evaluation, prompt engineering, and pipeline development. Maintain comprehensive documentation of pipeline architectures, evaluation methodologies, and experimental results.

• 5 - 8 years of experience in a Data Scientist, Computer Vision Engineer, or ML Engineer role.
• Bachelors degree in Data Science, Mathematics, Computer Science, Statistics, or a related field (Masters degree is an asset).
• Proven experience building and deploying computer vision or video analytics pipelines in production environments using Python.
• Hands-on experience with Vision Language Models (VLMs), large multi-modal models, and cloud-based inference infrastructure (e.g., Ray, Vertex AI, or Azure). Proficiency in writing SQL queries and working with large-scale data (BigQuery, Spark).
• Experience with evaluation framework design, including ensemble voting approaches and ground truth management.
• Excellent communication skills for effective collaboration across product, platform, and data science teams