Staff Software Engineer - Scalable AI Training Systems

River AI Inc.; Palo Alto (CA); On-site; USD 200,000 - 420,000; Full time

Benefits offered by this job

Health insurance

Unlimited PTO

Relocation assistance

Visa sponsorship

Job summary

River AI Inc. is seeking exceptional systems engineers to build high-performance engines that train our models. You will own the core infrastructure stack, from custom GPU kernel work to managing clusters of thousands of nodes, ensuring researchers can focus on science rather than bottlenecks.

You will collaborate with scientists to rapidly implement, optimize, and scale experimental model architectures, driving fast, reliable, and massively scalable training and inference workflows.

Qualifications

• Bachelor's degree in CS/CE or equivalent practical experience.
• Deep expertise in systems-level languages (C, C++, or Rust).
• Strong foundation in architecture, memory management and concurrency.
• Exceptional debugging skills in distributed environments.
• Collaborative mindset and bias for action.

Responsibilities

• Architect and deploy fault-tolerant distributed systems for training and inference across clusters.
• Design high-performance kernels to maximize tensor operation efficiency and memory throughput.
• Profile end-to-end to resolve blockers across hardware and software pipelines.
• Collaborate with researchers to scale experimental model architectures.

Skills

C/C++/Rust

Distributed systems

GPU kernels

Debugging

Education

Bachelor's degree in Computer Science/Engineering

Job description

River AI Inc. is seeking exceptional systems engineers to build high-performance engines that train our models. You will own the core infrastructure stack, from custom GPU kernel work to managing clusters of thousands of nodes, ensuring researchers can focus on science rather than bottlenecks.

You will collaborate with scientists to rapidly implement, optimize, and scale experimental model architectures, driving fast, reliable, and massively scalable training and inference workflows.