ML Model Performance Engineer - Inference and Acceleration

Baseten; New York (NY); On-site; USD 200,000 - 275,000; Full time

Benefits offered by this job

100% coverage of medical, dental, and vision insurance

Generous PTO policy

Paid parental leave

Company-facilitated 401(k)

Exposure to various ML startups

Job summary

A leading AI platform provider in New York is looking for a Software Engineer focused on ML performance. This role involves optimizing ML model inference using cutting-edge techniques and collaborating with a diverse team. Ideal candidates have a degree in Computer Science or related fields, along with experience in Python or C++. The position offers competitive compensation along with comprehensive benefits including equity, health insurance, and generous PTO.

Qualifications

• Bachelor’s, Master’s, or Ph.D. degree in Computer Science, Engineering, Mathematics, or related field.
• Experience with programming languages, such as Python or C++.
• Strong familiarity with ML libraries, especially PyTorch and TensorRT.

Responsibilities

• Implement and productionize cutting-edge techniques for ML model inference.
• Debug ML performance issues in underlying codebases.
• Collaborate with a diverse team to design and implement innovative solutions.

Skills

Experience with Python

Experience with C++

Familiarity with LLM optimization techniques

Strong familiarity with PyTorch

Understanding of GPU architecture

Education

Bachelor’s, Master’s, or Ph.D. degree in Computer Science or related field

Tools

Docker

Kubernetes

CUDA

Job description

A leading AI platform provider in New York is looking for a Software Engineer focused on ML performance. This role involves optimizing ML model inference using cutting-edge techniques and collaborating with a diverse team. Ideal candidates have a degree in Computer Science or related fields, along with experience in Python or C++. The position offers competitive compensation along with comprehensive benefits including equity, health insurance, and generous PTO.