Genesis AI
Inference
Bay Area · On-site · Full-time · Posted Apr 30, 2026
- Experience
- 8+ years
About the job
Build and optimize high-performance inference systems for robotics, spanning low-latency on-device deployment and distributed GPU-cluster serving.
What you'll do
- Build on-device inference pipelines for real-time next-token and diffusion-based robotics control loops.
- Design and optimize distributed GPU-cluster inference, including throughput, resource utilization, latency, and workload scheduling.
- Implement low-level performance code and develop monitoring and debugging tools for reliability and regression diagnosis.
What they ask for
- Deep experience in distributed systems, ML infrastructure, or high-performance serving.
- Production-grade expertise in Python and a strong background in systems languages.
- Low-level performance expertise in CUDA, Triton, kernel optimization, quantization, and memory and compute scheduling.
- Track record scaling inference workloads in throughput-oriented clusters and latency-critical on-device deployments.
Tools
Python, C++, Rust, Go, CUDA, Triton
AI focus
Inference for real-time robotics control, including next-token and diffusion-based workloads.
Model-written summary · the original posting is authoritative