Torc is developing software for automated trucks to transform how the world moves freight. The ML Frameworks Team is seeking a Software Engineer 2 to help build and maintain their machine learning training infrastructure, supporting the development and maintenance of ML training framework components for distributed model training in the cloud.
Requirements
- Good knowledge of Python and familiarity with PyTorch or other ML frameworks
- Experience with AWS services (S3, Lambda, EC2, or similar cloud platforms)
- Understanding of software engineering best practices including version control (Git) and agile workflows
- Experience working with data processing pipelines or APIs
- Previous experience with ML Ops workflows including dataset management, model training pipelines, and model registries
- Exposure to distributed computing concepts or frameworks like Ray
- Experience with PyTorch Lightning or similar training frameworks
Responsibilities
- Implement and maintain machine learning training pipelines and infrastructure
- Support the development of our Ray-based framework for distributed machine learning training and data transformation pipelines
- Help maintain the shared ML codebase used by all machine learning projects at Torc
- Collaborate with researchers and engineers to support their machine learning workflows
- Participate in building internal tools to enhance ML operations and developer productivity
- Write clean, well-documented code and participate in code reviews
- Learn and apply best practices for cloud-based ML infrastructure using AWS services
Other
- Bachelor's degree in computer science, data science, artificial intelligence or related field with 1+ years of professional experience.
- Strong problem-solving skills and ability to learn new technologies quickly
- Excellent collaboration skills and ability to work in a team environment
- Familiarity with AGILE development practices and tools such as Git and Jira
- Participate in the on-call rotation to support the health of our services and framework
