Sr. Staff ML Researcher, LLM Algorithmic Optimization

What you'll do

• Develop and evaluate numerical and model-optimization techniques for LLM inference, including quantization, compression, sparsity, pruning, distillation, and low-rank approximation.
• Optimize transformer workloads such as matrix operations, attention, and KV-cache utilization; assess tradeoffs among quality, precision, memory, and performance.
• Build Python research prototypes and collaborate with hardware, compiler, and software teams to map algorithms to accelerator capabilities.

What they ask for

• Master's degree or PhD in a quantitative field, with 5+ years of relevant hands-on experience.
• Strong machine-learning and applied-mathematics foundation, with hands-on experience developing and evaluating ML algorithms, models, or research prototypes.
• Strong Python programming skills, including object-oriented code design.
• Experience with modern ML frameworks such as PyTorch, JAX, or TensorFlow, and improving model performance, efficiency, or quality through algorithmic or numerical techniques.

Nice to have

• Familiarity with neural-network architectures; experience with transformers, LLM inference, or hardware-aware ML optimization is advantageous but not required.

Pay

$175K – $265K (base salary)

Tools

Python, PyTorch, JAX, TensorFlow

AI focus

Develops and optimizes methods for efficient LLM inference on DNN accelerators, including quantization, compression, sparsity, pruning, distillation, and hardware-aware execution.