Get Started
Hunting for you, 24x7Get the iPhone app

Your hunt runs in your pocket, and tells you the moment a strong match opens.

Download on theApp Store
LaddersPosted 6 days ago

Research Engineer - Inference

United StatesMid levelRemote or on-siteFull-time$150,000 – $350,000 a year
Who we'd reach at LaddersFound and reached by Power Apply
Best leadHiring manager

Leads the team this role joins

Founder

Decides on early hires

Head of talent

Runs hiring at the company

Power Apply

From your own Gmail and LinkedIn.

About the role

A recruiter is advertising a Research Engineer role focused on inference for a client in the information technology sector, open across the United States. The role builds high-performance inference systems for reinforcement learning and long-horizon agent workloads, and suits engineers with at least five years of GPU inference optimization experience. The listing is an excerpt, and the full description is published separately on LinkedIn.

What you’ll do

  • Extend inference support for new model architectures used in reinforcement learning training.
  • Raise rollout throughput for long-horizon agents by tuning concurrency.
  • Optimize multi-node mixture-of-experts inference, including expert parallelism.
  • Reduce the overhead of pushing weight updates from trainers to inference systems through sharding.
  • Evaluate lower-precision techniques and monitor for numerical regressions.
  • Work with model researchers to improve inference efficiency and end-to-end performance.

What they’re looking for

  • At least five years of GPU inference systems optimization experience.
  • Strong Python and PyTorch proficiency.
  • Experience profiling and debugging machine learning workloads.
  • Familiarity with distributed ML infrastructure.

Nice to have

  • Familiarity with vLLM or other inference engines.
  • Knowledge of CUDA or Triton kernel development for low-precision inference.

What’s on offer

  • Cash compensation of USD 150,000 to 350,000, including equity incentives.
  • Flexible arrangements, with remote or in-person options.
  • Visa sponsorship and relocation assistance for international candidates.
  • Quarterly team off-sites, hackathons and conference attendance.

Skills

Questions about this role

How much experience is needed?

At least five years in GPU inference systems optimization.

What is the pay range?

USD 150,000 to 350,000 in cash compensation, including equity incentives.

Is visa sponsorship offered?

The listing mentions visa sponsorship and relocation assistance for international candidates.

Is this the full job description?

No. The listing is an excerpt, and the complete description is on LinkedIn.