About SoftBank: SoftBank is making significant investments in infrastructure for AI. SoftBank Corp. has recently established a new US center in Sunnyvale CA, focused on infrastructure software for AI and AI foundations for mobile networks. As an AI/ML Engineering Intern, you will contribute to the research and development of AI inferencing frameworks in the Infrinia team which is building Inference as a Service.
You will work closely with a team of experts to push the boundaries of AI infrastructure and applications, helping to shape products that redefine AI for modern data center and edge use cases. We offer SOTA large-scale GPU infrastructure and high-energy collaborative environment for cutting-edge research & development.
This internship is suitable for graduate students. We will consider exceptional upper division
undergraduate students. Timing of these internships is flexible and some can also be made available round the year.
Qualifications:
• Enrolled in the graduate or upper division of bachelor’s degree in Computer Science, Electrical
Engineering, or a related field.
• Coursework and proficiency to support AI/ML understanding.
• Familiarity/Experience with AI frameworks such as PyTorch, TensorFlow, Dynamo, vLLM.
• Familiarity/Experience with LLMs, workflows and cutting-edge research.
• Hands-on projects demonstrating experience with model frameworks, inferencing, evaluation and benchmarking.
Responsibilities Include:
• Contribute to developing and optimizing AI frameworks (e.g., Dynamo, vLLM).
• Design and execute evaluations to measure model quality and efficiency.
• Deploy serving instances and benchmark inferencing performance.
• Optimize resource utilization on large-scale GPU clusters for AI workloads.
• Stay current with advancements in AI/ML, contribute to cutting-edge research, implement
findings to improve performance and author papers.
Salary and Benefits:
Competitive base salary, relocation. Great Sunnyvale downtown location with access to Caltrain,
restaurants and facilities.
