Senior Research Engineer

About This Opportunity

We are representing a confidential AI technology organization seeking a Senior Research Engineer to join their machine learning research team. Our client is a high-growth company working at the frontier of large-scale model development, serving hundreds of millions of inference requests monthly and processing over a million hours of data daily.

This role sits at the intersection of research and infrastructure. You'll be responsible for developing and improving the systems that enable large-scale distributed training, data processing, and production inference. The focus is on raising experimental velocity—making it faster for the research team to launch experiments, measure results, and iterate on what matters most.

Responsibilities

• Maintain and evolve the JAX training framework for large-scale distributed training on TPU infrastructure
• Increase experimental velocity by improving tooling, measurement systems, and the speed at which the team can validate hypotheses
• Investigate data quality issues and build systems that surface problems before they affect model accuracy
• Build evaluation harnesses and analyze production model accuracy to identify high-impact improvement opportunities
• Translate research prototypes into production-ready systems, refactoring and modernizing architectures as needed
• Optimize production inference systems for speech and language models using techniques such as quantization, speculative decoding, and advanced sharding strategies
• Resolve performance bottlenecks across the stack, from low-level kernels (XLA, Pallas) to high-level system design
• Collaborate with researchers, infrastructure teams, and production engineering to trace issues to root causes and ship durable fixes

What We're Looking For

• Expert-level proficiency with JAX and TPUs, including the surrounding ecosystem (Flax, Optax, XLA compilation pipeline)
• Strong measurement discipline: you define success metrics upfront, stay skeptical of results until validated, and investigate unexpected improvements as rigorously as failures
• Deep experience optimizing inference systems for production, ideally with large language models or speech models
• Proven background in distributed training at scale and modern ML infrastructure best practices
• Familiarity with modern inference optimization techniques: continuous batching, KV-cache management, sharding strategies, and quantization
• Willingness to work across the full pipeline—while your core strength may be in JAX and TPU performance, you're motivated to trace problems into data quality, evaluation, or other domains when necessary
• Strong Python skills; C++ or Rust experience for kernel-level work is a plus
• Excellent communication skills and a collaborative mindset, with the ability to explain complex technical tradeoffs clearly
• Enthusiasm for refactoring and improving existing systems to make them faster and more reliable
• Bonus: Domain knowledge in speech-to-text systems, ASR architectures, audio processing, or streaming inference

How We Work With You

Candidates apply directly through Get A Job.ai. Our talent team will conduct an initial screen to assess fit and answer any questions about the role. If there's mutual interest, we'll submit your profile to our client for their review. Please do not contact the client directly—all communication and interview coordination will be managed through our recruiting team to ensure a smooth and professional process.

Pay

Pay via Get A Job.ai: $270,000 - $310,000 USD

Equal Opportunity

Get A Job.ai is committed to creating an inclusive recruiting process. We welcome candidates from all backgrounds and encourage applications from individuals of any race, gender identity or expression, sexual orientation, religion, origin, ability, age, or veteran status.