Senior ML Systems Engineer (JAX/TPU)

AssemblyAI, Inc.; Boston (MA); On-site; USD 270,000 - 310,000; Full time

Job summary

AssemblyAI is seeking a Senior Research Engineer to advance large-scale distributed training, data processing, and inference. You’ll work with JAX/TPUs, improve training pipelines, and push production-ready ML systems forward across the research and infrastructure stack.

You will collaborate with researchers, infrastructure, and production engineering to ship impactful improvements, focusing on end-to-end problem solving and measurable results.

Qualifications

• Expert-level proficiency with JAX and TPUs.
• Measurment discipline and rigorous evaluation mindset.
• Strong Python skills and ability to ship production-ready code.
• Excellent communication and collaborative mindset to coordinate across teams.
• Willingness to refactor and improve existing systems.

Responsibilities

• Raise the team's experimental velocity and enable faster experiment launches.
• Maintain and evolve the JAX training framework for scalable TPU runs.
• Improve training data quality and surface issues with measurable gains.
• Analyze production model accuracy and build evaluation harnesses.
• Translate research prototypes into production-ready systems and modernize architectures.
• Optimize production inference for speech language models, including quantization and decoding techniques.
• Investigate performance bottlenecks from kernel to system level and ship fixes.
• Collaborate with researchers, infrastructure, and production engineering to trace issues to their source.

Skills

Measurement discipline

Python skills

Strong communication

Collaborative mindset

Tools

JAX

TPUs

Flax

Optax

XLA

C++

Rust

Job description

AssemblyAI is seeking a Senior Research Engineer to advance large-scale distributed training, data processing, and inference. You’ll work with JAX/TPUs, improve training pipelines, and push production-ready ML systems forward across the research and infrastructure stack.

You will collaborate with researchers, infrastructure, and production engineering to ship impactful improvements, focusing on end-to-end problem solving and measurable results.