Luma AI

Luma AI

Research Scientist / Engineer – Reinforcement Learning Infrastructure

Redwood City, CA

Sponsorship not specified$30k-$60kDetected 43 days ago
KubernetesPyTorchLLMsResearchCommunication

Stay score

odds of building a lasting career here

35Risky
Cap-exempt (no lottery)0
Sponsors this role74
Entry-level history0
PERM / green-card track0
Lottery odds (Level I)39
Fits your clock70

Thin sponsorship signal and lottery-bound (~15% per draw). A low-probability bet with your clock running. Prioritize cap-exempt roles and proven entry-level sponsors first.

Lottery odds assume a STEM candidate.

Personalize to your clock →

H-1B wage level

the lottery is wage-weighted — each level is one more entry

Level I · 1×
Level I$100,4221 entry
Level II$137,7792 entries
Level III$175,1363 entries
Level IV$212,4934 entries

The bottom of this range ($30,000) is below the Level I prevailing wage of $100,422. An H-1B cannot be filed below the prevailing wage, so an offer at the floor of this band could not be sponsored as posted.

DOL prevailing wage, 2026-27 wage year · Computer and Information Research Scientists (15-1221) · San Francisco-Oakland-Fremont, CA. Wage level is derived by USCIS from the offered wage, occupation and worksite; the occupation shown is inferred from the job title.

Employer immigration record

from this employer's Department of Labor filings

Files H-1B transfers

16 transfer filings in the last year, covering 16 workers. Median labor-condition decision: 7 days. An employer that already files transfers is one that can take over an existing H-1B.

Sourced from Department of Labor LCA, PERM and prevailing-wage disclosure data. Employer matching is by name, so figures may be split across an employer's legal entities. Absence of a filing means none appears in our copy of the data, not that none exists.

Community outcomes

No reports yet — be the first to help the next applicant.

About the role

  • RL is how Luma's models go from capable to useful.
  • RL at scale is a full-loop systems problem: training, rollout generation, environment execution, and reward computation running concurrently across thousands of GPUs, all needing to stay fast, stable, and correct together.
  • It fits someone who has lived this - post-trained LLMs with RL, built environments and verifiers, and debugged asynchronous rollout pipelines at scale.

Requirements

  • Strong understanding of GPU clusters, networking, and communication libraries (NCCL, MPI) under mixed training and inference workloads.
  • Hands-on experience post-training LLMs with RL (PPO/GRPO-family, RLHF, RLVR) at meaningful scale.
  • Extensive distributed PyTorch training and parallelism (FSDP, Tensor/Pipeline/Expert Parallel) for foundation models.
  • Experience building RL environments, reward functions, verifiers, or evaluation harnesses for LLM agents, including sandboxed execution and multi-turn tool use.
  • Deep familiarity with RL post-training frameworks (veRL, OpenRLHF, TRL, Ray orchestration) and rollout inference engines (vLLM, SGLang).

Nice to have

  • Running RL training across 100+ GPUs, including asynchronous or disaggregated trainer/rollout architectures.
  • Containerization and orchestration (Kubernetes, Ray) for large environment fleets and sandboxed workloads.
  • Research contributions in RL for LLMs, or open-source contributions to RL training frameworks.
  • Luma is an equal opportunity employer.

Compensation

  • $30k-$60k

Benefits

  • We believe multimodality is critical for intelligence - the next step beyond language models comes from vision.

Company info

  • Luma's mission is to build unified general intelligence that can generate, understand, and operate in the physical world.

This listing is sourced directly from Luma AI's careers page and normalized into a canonical job model.