ML Infrastructure Engineer, AI Research Team

As an ML Infrastructure Engineer at Gatik, your main responsibilities will include owning and leading the exploration of distributed training optimization, building scalable training and inference pipelines, and collaborating closely with the AI Research and DevOps teams. You'll also develop benchmarking tools, augment data pipelines, and adjust frameworks to enhance machine learning development.

To be considered for the ML Infrastructure Engineer role at Gatik, you should have a minimum of 3 years of production or research experience in ML Infrastructure, specifically in distributed training and model inference. A strong foundation in programming, knowledge of deep learning algorithms, and familiarity with cloud products like Azure, AWS, or GCP are also essential.

The ML Infrastructure Engineer at Gatik works closely with the AI Research team and DevOps team to prepare the necessary assets and tools for machine learning development. This collaboration ensures that the infrastructure you build optimizes and supports ongoing research in autonomous driving applications.

Candidates should be proficient in programming languages such as Python and C++, and have experience with ML frameworks like TensorFlow, PyTorch, or JAX. Knowledge of Kubernetes clusters, distributed compute, and GPU programming techniques is also vital for this role at Gatik.

Gatik prides itself on fostering a diverse and inclusive work environment that encourages collaboration, respect, and innovation. As an ML Infrastructure Engineer, you'll join a team of talented individuals who share a passion for technology and autonomous driving, all while feeling supported and empowered in your contributions.

In answering this question, elaborate on specific projects where you implemented distributed training techniques, such as model parallelization or data parallelism, to enhance model performance and reduce training time. Provide examples that highlight your understanding of sensitive parameters and trade-offs involved.

When discussing your experience, mention specific cloud platforms like Azure, AWS, or GCP that you have used to implement and manage ML workflows. Talk about challenges faced, tools utilized (e.g., managed services like SageMaker or Azure ML), and how you achieved scalability and efficiency.

Highlight the importance of open communication and regular check-ins with AI Research and DevOps teams. Share how you ensure alignment with their goals, such as setting milestones and leveraging tools like project management software to track progress and facilitate discussions.

Focus on your systematic approach to debugging, which may include logging important metrics, monitoring resource utilization, and using visualization tools. Explain how you troubleshoot specific issues that arise during model training and inference in distributed environments.

Discuss any experience you have in developing or using model benchmarking tools. Provide details about frameworks or libraries utilized for benchmarking and how you analyze results to improve model performance and reduce latency.

Emphasize the writing of clean, well-documented code and comprehensive testing procedures. Talk about maintaining coding standards, regular audits of algorithms, and leveraging continuous integration/continuous deployment (CI/CD) pipelines for automated testing.

Share your hands-on experience with frameworks like TensorFlow, PyTorch, and JAX, explaining your preference for certain frameworks based on the type of project, ease of use, community support, or performance metrics.

Describe how you've used Kubernetes to manage containerized workloads for machine learning tasks, including deploying, scaling, and maintaining models in a production environment. Highlight use cases that demonstrate the effectiveness of Kubernetes in orchestrating ML workloads.

Talk about how you keep yourself updated by participating in workshops, online courses, attending conferences or webinars, and engaging with the ML community through forums and publications to stay informed about latest technologies and best practices.

Share your passion for autonomous driving and how it aligns with your skills and experiences. Discuss the potential for innovation and societal impact that autonomous technology can bring, as well as your personal motivation for contributing to this exciting field.