Senior Software Engineer - ML Performance at Latitude AI
Pittsburgh, PA
$179,200–$268,800from the description
Aug 2, 2026
Pittsburgh, PA
Sep 24, 2026
What this job asks for AI summary
A performance engineering role on Latitude AI's autonomous vehicle software stack, focused on optimizing AI/ML inference pipelines from prototype to production on embedded, heterogeneous compute hardware. The work spans profiling and tuning PyTorch models through to C++ runtime code, building evaluation infrastructure, and collaborating cross-functionally to squeeze maximum performance from on-vehicle compute resources. Suits engineers with a strong background in GPU/DSP architecture, deep learning deployment, and performance-critical systems programming.
Senior level · 4+ years · Pittsburgh, PA · Bachelor's required · Full-time
Posted 2 times — it's one opening, so apply once.
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 60 people in the Pittsburgh, PA area plausibly meet what this posting asks for (data scientists). range 25–95
Applicant volume Moderate — A normal amount of company. The rare requirements below are what will separate a shortlisted application from the rest.
Most people in this occupation already list these. Still required — just not what gets you shortlisted.
What the occupation pays Median $98,796 (middle half $74,401–$130,048). 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 Aug 4, 2026. It is a model, not a headcount.
Why we read it this way (7)
The salary range ($179,200–$268,800/yr) is explicitly stated for California; the role is headquartered in Pittsburgh with additional engineering centers in Dearborn, MI and Palo Alto, CA — compensation may differ by location.
Total experience minimum is 4+ years with a Bachelor's degree; the JD offers equivalents (Master's + 2 years, or PhD with no stated minimum years), so 4 years is used as the floor.
GPU/DSP architecture and heterogeneous embedded computing are listed together as a single required expertise; GPU architecture is captured as the primary skill name.
Model compression techniques (quantization, pruning, kernel fusion, distillation) and SIMD coding appear under the 'What will make you stand out' section and are treated as preferred.
Embedded systems experience is inferred from the 'stand out' section's reference to production embedded system profiling and deployment flows — listed as preferred.
Visa sponsorship is noted as available; candidates must be authorized to work permanently in the US.
Ignored 1 non-technology phrase(s) as skills (responsibilities/concepts, not named tools): GPU architecture.
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.