Senior AI Stack Engineer, FlexAI

• The ideal candidate will have significant experience in software engineering for large-scale distributed systems, with a passion for enhancing AI infrastructure and ensuring seamless model training and deployment processes
• 5+ years of experience in software engineering, with a focus on runtime systems or performance optimization for large-scale distributed systems
• Strong expertise in low-level performance optimizations and systems programming (C/C++, Go, etc.), with Python experience preferred
• A Bachelor’s or Master’s degree in Computer Science, Software Engineering, or a related technical field
• Proven experience working with distributed systems, and the ability to optimize runtime environments for AI workloads
• Excellent problem-solving skills with the ability to innovate and think outside the box in a fast-paced, evolving environment
• Strong communication and collaboration skills, comfortable working cross-functionally with engineering, research, and product teams

Desirable

• FlexAI is seeking a talented and driven AI Stack Engineer to join our innovative team. In this role, you will be responsible for improving the reliability, scalability, and performance of our custom AI software stack, with a focus on PyTorch optimization and model deployment
• Improve LLM Training Reliability and Elasticity:
• Design and implement solutions to make PyTorch more elastic and resilient, enabling fault tolerance and dynamic scaling of training jobs
• Collaborate with teams to enhance PyTorch functionalities and reduce training downtime, optimizing large language model (LLM) training workflows
• Optimize Model Packaging and Production Runtime:
• Ensure seamless integration of customer models with our custom PyTorch stack
• Manage and improve the production runner, focusing on performance, scalability, and deployment processes to ensure efficient model execution
• Develop and Maintain Internal Tools for Model Training:
• Contribute to the development of internal libraries and tools to improve the training process, including implementing asynchronous operations and fault recovery mechanisms
• Maintain code quality, enforce best practices, and ensure continuous integration for production readiness