Senior Machine Learning Engineer, ML Efficiency
$216,700–$303,400
Jul 23, 2026
—
Jul 25, 2026
What this job asks for AI summary
A senior ML engineering role focused on making Ads machine learning systems faster and more cost-efficient. The work spans training pipelines, inference and serving paths, and launch-readiness tooling — diagnosing production bottlenecks through profiling and benchmarking, building reusable optimization primitives, and collaborating with ranking, serving, and platform teams. Suits engineers with hands-on experience improving real production ML workloads at both the model and systems level.
Senior level · Remote · Full-time
“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
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).
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 (6)
This role is remote-eligible in any country where Reddit has a physical office presence — it is not strictly US-only, though Reddit's primary presence is in the US. The location field is left blank because no specific metro is required.
The SOC classification is a judgement call: the role is primarily about building performance tooling, optimization primitives, and efficiency infrastructure for ML systems (pointing to 15-1252 Software Developers), but it sits very close to ML modeling work (15-2051 Data Scientists). 15-1252 was chosen because the dominant output is engineered systems and tooling, not statistical modeling or research.
No overall years-of-experience figure is stated in the posting; seniority is inferred from the 'key senior engineer' framing and the scope of independent ownership described.
All Nice-to-have items (PyTorch, distributed training, GPU migrations, quantization/pruning/distillation, load testing, cost observability) are listed under an explicit 'Nice-to-have' heading and are marked as preferred accordingly.
The required skills are described in qualitative terms ('deep experience', 'direct hands-on experience', 'strong technical judgment') with no specific named technologies mandated in the requirements section — the hard gates are capability-level, not tool-level. Named tools appear only in the Nice-to-have section.
Ignored 2 non-technology phrase(s) as skills (responsibilities/concepts, not named tools): model training optimization, model serving/inference optimization.
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.