Senior Computer Vision Engineer

About the role

Join the Algorithms division at Sword Health, where you will play a pivotal role in enhancing the "Thrive" initiative, which utilizes AI to provide guided physical therapy for individuals dealing with chronic pain. This position is centered around the development and implementation of computer vision models that analyze human movement through camera feeds. Your contributions will be vital in transforming real-time motion data into actionable clinical insights for our users.

Key facts

What you'll do

• Take charge of essential computer vision models, such as 3D human pose estimation and statistical body modeling, overseeing their journey from initial concept through to production deployment.
• Optimize these models to function in real-time, both in cloud environments and on mobile devices like tablets.
• Oversee the complete lifecycle of data and code, which includes managing training frameworks, annotation processes, pipelines, auto-labeling, test sets, and data taxonomy.
• Enhance the understanding of movement by employing multimodal strategies that integrate vision with language and reasoning, leading to innovative solutions in movement intelligence.
• Standardize and refine processes for model training, tracking, versioning, and deployment, ensuring that all results are reproducible and testable.
• Create automated systems that enable the team to scale its output beyond manual efforts.
• Foster the growth of the Computer Vision team by establishing and advocating for best practices and guiding principles within the organization.

Requirements

• A minimum of 5 years of experience in addressing intricate computer vision challenges, with a proven track record of deploying models in a production environment.
• A robust background in software engineering, encompassing architecture, pipelines, MLOps, and the entire model lifecycle.
• In-depth, hands-on experience with contemporary computer vision methodologies, including transformers and convolutional networks, with specialized skills in detection, segmentation, tracking, pose estimation, or 3D modeling.
• A data-centric mindset, evidenced by the creation of custom datasets, development of data flywheels, implementation of active-learning loops, and treating datasets as if they were source code.
• A strong grasp of multimodal AI, with the ability to construct systems that merge vision with language and reasoning as required.
• Exceptional written and asynchronous communication skills, with a commitment to documenting your thoughts and designs for the benefit of others.
• Demonstrated ability to work independently and take full ownership of projects, capable of defining plans in the face of ambiguity and delivering results autonomously.
• Proficiency in PyTorch or JAX, familiarity with the Python data ecosystem, and a willingness to learn new programming languages and codebases.
• Experience with AI-native workflows, integrating AI into everyday processes, including agents and LLM-in-the-loop tools.

Nice to have

• Familiarity with pose estimation, body modeling, movement intelligence, or human motion analysis.
• Experience in handling large datasets and constructing full-lifecycle machine learning systems within a startup or scaleup context, taking on a variety of responsibilities.
• Background in automating experimentation or cycles for model improvement.
• A product-oriented approach, empathy for users, and a strong desire to create a meaningful impact for millions of users.

Skills & tools

• Computer Vision
• 3D Human Pose Estimation
• Statistical Body Modeling
• PyTorch
• JAX
• Python Data Stack
• MLOps
• Transformers
• Convolutional Models
• Detection
• Segmentation
• Tracking
• Multimodal AI
• Language Models
• Reasoning

Practical notes

Candidates must hold a valid EU visa and be currently residing in Portugal, as relocation assistance is not available. Benefits include comprehensive health, dental, and vision insurance, a meal allowance, equity shares, a remote work allowance, flexible working hours, options for remote work, discretionary vacation days, and office snacks and beverages. All applicants will be evaluated for their proficiency in AI, with a minimum requirement of Level 1 (Explorer) and an expectation of Level 3 (Integrator) for this position.