24-MAG · New York, NYremote

Salary
$65–$105/hrfrom the description
Posted
Jul 19, 2026
Location
New York, NY
Last confirmed open
Jul 20, 2026

What this job asks for AI summary

A full-time remote contract role for experienced MLOps and ML systems engineers to support a generative AI training initiative. Day-to-day work involves designing technically rigorous tasks and solutions covering distributed training, ML framework internals, and custom GPU kernel development in Pallas or Triton, then evaluating model-generated outputs against those standards. Suits engineers with production-scale JAX or PyTorch experience and a background in GPU performance optimization.

Mid level · 2+ years · Remote · Contract

Must have (4)
JAX or PyTorchPallas or Tritondistributed trainingMLOps
Nice to have (3)
XLACUDALLM

“or” means any one of them counts — you don't need all of them.

We read this from the posting text with AI. Skim the description below before ruling yourself out.

How this req sits in the market our data

Roughly 3,150 people nationally plausibly meet what this posting asks for (software developers). range 1,300–6,800

Applicant volume Light — Few people clear these requirements, so an application that does clear them gets looked at. Worth applying to even if you miss a nice-to-have.

What gives you an edge
Pallas2%Triton2%MLOps8%JAX9%PyTorch9%

Rare in this occupation — lead with these, and say what you built with them.

What the occupation pays Median $138,970 (middle half $107,524–$175,762). This posting is about at that midpoint.

Estimated from BLS employment for this occupation and area, per-skill prevalence across our listing corpus, and published wage benchmarks — as of Jul 28, 2026. It is a model, not a headcount.

Why we read it this way (10)

JAX and PyTorch are listed together as a paired requirement ('JAX, PyTorch, or both') — each is emitted separately as a hard gate since the posting firmly requires production experience with at least one; alternatives are cross-listed accordingly.

Pallas and Triton are similarly paired ('Pallas or Triton') as the required GPU kernel tooling — emitted separately with each as the other's alternative.

The posting is described as a W-2 contingent (contract) engagement, not a permanent hire, despite the 'full-time' framing.

Seniority is assessed as Mid: the minimum stated experience is 2 years and the role is framed around task design and evaluation work rather than broad architectural ownership, despite the advanced technical subject matter.

Degree is marked None: the JD states a relevant degree is 'highly relevant' and graduate education 'may be helpful', but explicitly accepts equivalent professional experience — no hard degree gate.

XLA, CUDA, LLM/generative AI experience, and GPU benchmarking appear under the 'Nice to Have' section and are marked preferred.

SOC confidence is Medium: the role blends ML infrastructure engineering (15-1252) with evaluation/annotation of model outputs, which has no clean SOC mapping; the primary day-to-day work described (designing tasks, writing solutions, reviewing GPU kernels, assessing training pipelines) most closely aligns with Software Developers.

Ignored 1 non-technology phrase(s) as skills (responsibilities/concepts, not named tools): GPU kernel development.

Posting is for a contract engagement — the market benchmarks below price full-time roles, so read the comp comparison with that in mind.

Caller marked this a fully-remote role — scored against the national candidate pool.

Read the full posting

The employer publishes the full description on their own site — read it there ↗. Or sign in to read it here — it's free, and it also lets you track this application.

Apply

Apply on employer site ↗