Senior AI Performance Engineer

What you'll do

• Bring up and optimize foundation models on the SambaNova platform.
• Profile performance across compiler, runtime, and hardware layers and identify bottlenecks.
• Collaborate across teams to develop scalable inference solutions and apply model, quantization, scheduling, and memory optimizations.

What they ask for

• Bachelor’s or higher degree in computer science, electrical engineering, or a related field.
• 3+ years of experience in one or more listed areas: deep learning optimization, compiler/runtime/kernel optimization, or software–hardware co-design and systems performance tuning.
• Proficiency in Python or C++, with strong foundations in algorithms, data structures, and numerical computing.
• Experience with at least one of PyTorch, TensorFlow, or JAX, and ability to analyze and optimize real-world ML pipeline performance.

Nice to have

• Hands-on experience with LLM or multimodal model training and inference; distributed training or high-throughput inference systems.
• Familiarity with quantization, graph optimization, kernel fusion, model partitioning, or frameworks such as DeepSpeed, Megatron, vLLM, or TensorRT.

Pay

$180,000 — $255,000 USD (base pay range)

Tools

Python, C++, PyTorch, TensorFlow, JAX, DeepSpeed, Megatron, vLLM

AI focus

Optimizes large-scale AI inference, including bringing up foundation models and improving their throughput, latency, and efficiency on SambaNova hardware.