Uber
Senior Software Engineer – ML Infra
$202K–$224K
Sunnyvale, California +2 · Senior · Full-time · Posted Aug 13, 2026
- Location
- Sunnyvale, California
San Francisco, California
Seattle, Washington - Experience
- 6+ years; experience leading large-scale engineering projects from inception through global productionization is also required, with no duration stated.
- Education
- Bachelor’s degree in Computer Science, Engineering, Mathematics, or equivalent practical experience.
About the job
Build and lead Uber’s machine-learning training infrastructure, scaling systems for large datasets and models while improving performance and resource use.
What you'll do
- Design and build end-to-end systems for the ML training stack; solve scaling challenges for large datasets and models.
- Mentor engineers and guide architecture, code reviews, and large-scale projects through global productionization.
- Develop tools and use statistical analysis to diagnose performance regressions and optimize infrastructure resource utilization.
What they ask for
- 6+ years of professional experience building scalable, fault-tolerant products or platforms.
- Proficiency in one or more programming languages, such as Go, C/C++, Java, Python, or JavaScript.
- Proficiency in deep-learning technologies, such as distributed training frameworks, LLMs, or GPU optimizations.
- Bachelor’s degree in Computer Science, Engineering, Mathematics, or equivalent practical experience.
Nice to have
- Advanced degree (Master’s or PhD) focused on distributed computing, system performance, or machine learning.
- Experience contributing to open-source projects or leading industry-level technical discussions.
Pay
USD $202,000 per year - USD $224,000 per year (base salary)
Tools
Go, C/C++, Java, Python, JavaScript, distributed training frameworks, LLMs
AI focus
Build and scale ML training infrastructure, including systems for large models and datasets; work with distributed training frameworks, LLMs, and GPU optimizations.
Model-written summary · the original posting is authoritative