Drimva

Uber

Staff Machine Learning Engineer - Uber AI Solutions

$232K–$258K

San Francisco, California +1 · Staff · Full-time · Posted Oct 23, 2026

Location
San Francisco, California
Sunnyvale, California
Experience
5+ years of professional experience; no alternative stated.
Education
Bachelor's or equivalent work experience in Computer Science, Machine Learning, Engineering, Mathematics, or a related discipline.

About the job

Build and lead end-to-end machine learning and backend products for AI data and evaluation, owning customer outcomes and business metrics.

What you'll do

  • Build and operate ML and backend systems that become customer-facing products.
  • Own a product vertical from identifying opportunities and setting direction through delivery and business results.
  • Partner with customers, product managers, and engineers to grow accounts and raise execution quality.

What they ask for

  • Bachelor's or equivalent work experience in Computer Science, Machine Learning, Engineering, Mathematics, or a related discipline.
  • 5+ years designing, building, and operating production ML systems, including backend engineering from data to serving.
  • Track record of owning a product or product line and driving business impact, such as revenue, adoption, or customer value.
  • Fluency in Python and a production systems language such as Go, Java, or C++, and ability to build and operate reliable services.

Nice to have

  • Master's or PhD in a related discipline.
  • Experience growing a product or business line in a startup-like environment; applied ML in data or evaluation domains; production GenAI/LLM systems; or expertise in UAIS domains.

Pay

USD $232,000 per year - USD $258,000 per year (base salary; San Francisco, CA); USD $232,000 per year - USD $258,000 per year (base salary; Sunnyvale, CA)

Tools

Python, Go, Java, C++, GenAI, LLM, ASR, LiDAR

AI focus

Develop ML systems and data/evaluation infrastructure across modalities; preferred work includes production GenAI/LLM systems, model evaluation and tuning, ASR, agentic evaluation, and physical-world or multimodal data.

Model-written summary · the original posting is authoritative

First seen Oct 15, 2026 · Last verified Nov 5, 2026 (UTC)