ML Compiler Engineer for Custom ML Accelerators

Amazon Web Services (AWS); Seattle (WA); On-site; USD 180,000 - 240,000; Full time

Job summary

Amazon Web Services (AWS) is seeking a skilled ML compiler engineer to help build a state-of-the-art deep learning compiler stack that optimizes models for PyTorch, TensorFlow, and JAX across accelerators like Inferentia/Trainium.

You will design, implement, and optimize compiler features, collaborate with Runtime, Frameworks, and Hardware teams, and contribute to the Neural Compiler effort enabling scalable AI in the cloud.

Qualifications

• Experience designing, implementing, and optimizing compiler features for DL workloads.
• Collaborate across Runtime, Frameworks, and Hardware teams to optimize performance.
• Familiarity with ML frameworks (PyTorch, TensorFlow, JAX) and accelerators such as Inferentia/Trainium.

Responsibilities

• Design, implement, test, deploy and maintain compiler solutions to improve performance, durability, cost, and security.
• Research implementations that deliver the best possible experiences for customers.
• Design and code across the stack: scheduling, memory, data transfer, graph partitioning, code generation, ISA, and hardware bring-up.

Skills

Compiler engineering

C/C++

Performance optimization

Graph partitioning

Instruction scheduling

Memory management

Parallel programming

Cross-functional collaboration

Job description

Amazon Web Services (AWS) is seeking a skilled ML compiler engineer to help build a state-of-the-art deep learning compiler stack that optimizes models for PyTorch, TensorFlow, and JAX across accelerators like Inferentia/Trainium.

You will design, implement, and optimize compiler features, collaborate with Runtime, Frameworks, and Hardware teams, and contribute to the Neural Compiler effort enabling scalable AI in the cloud.