ML Engineer, AI Inference Platform - Hybrid Seattle

ElastixAI; Seattle (WA); Hybrid; USD 140,000 - 210,000; Full time

Benefits offered by this job

Equity package

Medical, dental, vision coverage

Flexible Time Off

Parental leave

Gym benefit

Learning & development stipend

Job summary

ElastixAI in Seattle, WA is seeking a Machine Learning Engineer to build our core AI inference platform. You will design and develop ML deployment components, optimization pipelines, and benchmarking workflows in a fast‑paced startup setting.

You will collaborate with a cross‑functional team to ensure our stack is scalable, reliable, and easy to use, while exploring cutting‑edge ML techniques across diverse hardware targets.

Qualifications

• PhD/MS in Computer Science, Computer Engineering, or a related field.
• 3+ years of machine learning R&D and deployment experience.
• Proficiency in Python or C++.
• Experience with PyTorch, TensorFlow, or JAX.
• Strong CS fundamentals, data structures, algorithms, and testing.
• Excellent problem solving and communication skills.
• Thrives in a fast-paced startup environment.

Responsibilities

• Design, develop, and maintain core components of our ML platform for scalability and reliability.
• Research and implement advanced ML techniques to optimize inference across hardware targets.
• Collaborate with systems and cloud engineers for efficient resource utilization.
• Contribute to API and tool design for seamless integration and management.

Skills

Python

C++

Strong communication

Problem solving

Education

PhD/MS in Computer Science/Engineering

Tools

PyTorch

TensorFlow

JAX

Job description

ElastixAI in Seattle, WA is seeking a Machine Learning Engineer to build our core AI inference platform. You will design and develop ML deployment components, optimization pipelines, and benchmarking workflows in a fast‑paced startup setting.

You will collaborate with a cross‑functional team to ensure our stack is scalable, reliable, and easy to use, while exploring cutting‑edge ML techniques across diverse hardware targets.