AI Performance Engineer: Scale ML Throughput & Training

Applied Intuition; Sunnyvale (CA); On-site; USD 180,000 - 260,000; Full time

Job summary

Applied Intuition, Inc. is seeking a performance engineer to optimize large-scale ML workloads in the datacenter.

You will own the gap between theoretical accelerator performance and real workloads, profiling across the stack to reduce wasted GPU-hours and improve throughput for training and offline inference. You will work at the intersection of accelerators, ML frameworks, and large-scale data infrastructure, partnering with teams to land performance wins that reduce time-to-result and cost

Qualifications

• Hands-on ML performance engineering experience with profiling and throughput optimization.
• Experience with distributed multi-node training at scale (e.g., FSDP, DeepSpeed, Megatron, NCCL).
• Deep familiarity with GPU/accelerator performance concepts: memory bandwidth, kernel overhead, occupancy.
• Experience with high-throughput or batch inference systems (e.g., NVIDIA Triton, TensorRT, ONNX Runtime).
• Fluency in Python and proficiency in C++ or another systems language.
• Excellent debugging, analytical, and problem-solving skills.
• Strong understanding of ML foundations and ability to solve non-playbook problems.

Responsibilities

• Profile and optimize distributed training end-to-end: data loading, preprocessing, kernel execution, gradient communication, checkpointing.
• Optimize large-scale offline and batch inference over petabyte-scale logs: batching, quantization, scheduling, saturation.
• Establish roofline and performance models; rank optimization opportunities by impact and effort.
• Improve multi-node scaling efficiency: sharding, parallelism, interconnect utilization, memory bandwidth, kernel fusion.
• Drive cluster goodput by reducing GPU idle time from input pipelines, I/O, scheduling gaps, and retries.
• Build benchmarking, observability, and regression-detection tooling to prevent silent degradation.
• Collaborate with engineers across functions to solve complex data and compute problems at scale.
• Contribute to a culture of collaboration, technical excellence, and innovation.

Skills

ML performance engineering

Distributed multi-node training

GPU/accelerator concepts

Batch inference systems

Python & C++ proficiency

Debugging & problem solving

ML foundations knowledge

Job description

Applied Intuition, Inc. is seeking a performance engineer to optimize large-scale ML workloads in the datacenter.

You will own the gap between theoretical accelerator performance and real workloads, profiling across the stack to reduce wasted GPU-hours and improve throughput for training and offline inference. You will work at the intersection of accelerators, ML frameworks, and large-scale data infrastructure, partnering with teams to land performance wins that reduce time-to-result and cost