Senior AI Inference Performance Engineer

SambaNova; San Jose (CA); On-site; USD 180,000 - 240,000; Full time

Job summary

SambaNova is seeking an ML performance engineer to optimize and scale state-of-the-art foundation models on its reconfigurable dataflow platform. You will work hands-on with models like DeepSeek, GPT OSS, and others to push throughput, latency, and efficiency while coordinating with compiler, runtime, and hardware teams.

Ideal candidates have strong ML performance tuning background, proficiency in Python or C++, and experience with major ML frameworks.

Qualifications

• Bachelor's or higher degree in computer science, electrical engineering, or a related field.
• 3+ years of experience in deep learning model development and performance optimization.
• Experience with compiler, runtime, or kernel-level optimization.
• Software–hardware co-design or systems performance tuning.
• Proficiency in Python or C++, with strong foundations in algorithms and numerical computing.
• Experience with PyTorch, TensorFlow, or JAX.
• Ability to analyze and optimize performance in real-world ML pipelines.

Responsibilities

• Bring up and optimize foundation models on the SambaNova platform through the software stack.
• Profile and enhance model performance across compiler, runtime, and hardware layers for high throughput and low latency.
• Collaborate with ML, compiler, runtime, and hardware teams to deliver high-performance AI applications.
• Integrate advances in model architecture, quantization, scheduling, and memory optimization.
• Develop robust, scalable end-to-end inference solutions aligned with customer needs.
• Identify bottlenecks and propose dataflow or scheduling optimizations for single-node and distributed systems.

Skills

Deep learning optimization

Python

C++

Algorithms/data structures

Perf analysis

Education

Bachelor's or higher in CS/EE

Tools

PyTorch

TensorFlow

JAX

CUDA

Triton

DeepSpeed

Megatron

vLLM

TensorRT

Job description

SambaNova is seeking an ML performance engineer to optimize and scale state-of-the-art foundation models on its reconfigurable dataflow platform. You will work hands-on with models like DeepSeek, GPT OSS, and others to push throughput, latency, and efficiency while coordinating with compiler, runtime, and hardware teams.

Ideal candidates have strong ML performance tuning background, proficiency in Python or C++, and experience with major ML frameworks.