TPU Systems Engineer — High-Performance ML Inference Equity

RadixArk; Palo Alto (CA); On-site; USD 180,000 - 250,000; Full time

Benefits offered by this job

Significant founding team equity

Comprehensive health benefits

Flexible work arrangements

Job summary

A leading AI infrastructure firm is seeking a TPU Systems Engineer to develop high-performance systems using JAX, XLA, and Pallas. This role involves pushing large-model workloads on TPU hardware and optimizing performance across the stack. Candidates should have at least 3 years of experience in production ML systems and a strong foundation in Python. Competitive compensation includes equity and comprehensive health benefits, with a US base salary range of $180,000 - $250,000.

Qualifications

• 3+ years experience building production ML systems with TPU-focused frameworks.
• Deep understanding of JAX/XLA internals: HLO, fusion, partitioning strategies.
• Experience with distributed systems like SGLang and training frameworks.

Responsibilities

• Build high-performance inference and training systems using JAX/XLA/Pallas.
• Optimize latency and throughput for LLM serving on TPU infrastructure.
• Collaborate with kernel engineers to achieve performance improvements.

Skills

Building production ML systems with JAX, XLA

Performance tuning across compiler and runtime layers

Proficiency in Python

Distributed inference systems

Kernel development skills

Education

Bachelor's or Master's degree in Computer Science or Electrical Engineering

Job description

A leading AI infrastructure firm is seeking a TPU Systems Engineer to develop high-performance systems using JAX, XLA, and Pallas. This role involves pushing large-model workloads on TPU hardware and optimizing performance across the stack. Candidates should have at least 3 years of experience in production ML systems and a strong foundation in Python. Competitive compensation includes equity and comprehensive health benefits, with a US base salary range of $180,000 - $250,000.