AI Performance Engineer

Join Applied, a leading company in the field of machine learning and autonomy. As an AI Performance Engineer, you will be responsible for optimizing large-scale machine learning workloads, focusing on distributed training runs and high-throughput batch inference. You will work at the intersection of accelerators, ML frameworks, and large-scale data infrastructure, collaborating with various teams to improve training time-to-result and offline processing cost. This role offers a competitive salary, comprehensive benefits, and opportunities for professional growth and development.

• Profiling and optimizing distributed training end to end, including data loading, preprocessing, augmentation, kernel execution, gradient communication, and checkpointing.
• Establishing roofline and performance models for workloads, quantifying the gap between achieved and theoretical performance, and stack-ranking optimization opportunities.
• Driving cluster goodput by reducing GPU idle time from input pipeline stalls, storage and network I/O, scheduling gaps, stragglers, and failure recovery on long-running jobs.

• Experience with high-throughput or batch inference systems (NVIDIA Triton Inference Server, TensorRT, ONNX Runtime, Ray, or similar)
• Fluency in Python and proficiency in C++ or another systems language
• Hands-on ML performance engineering experience: profiling, roofline analysis, throughput optimization, and root-cause investigation in production systems
• Deep familiarity with GPU or accelerator performance concepts - memory bandwidth, kernel launch overhead, occupancy, quantization, collective communication
• Excellent debugging, analytical, and problem-solving skills
• Experience with distributed multi-node training at scale (FSDP, DeepSpeed, Megatron, NCCL, or equivalent), including diagnosing scaling inefficiency as node count grows
• A deep understanding of machine learning foundations, and the ability to develop technical solutions for problems with no established playbook
• Experience with profiling toolchains such as Nsight Systems/Compute, PyTorch Profiler, or perf
• GPU kernel development experience: CUDA, Triton, CUTLASS, or hand-tuned attention implementations
• Experience with GPU scheduling and orchestration on Kubernetes, Slurm, or Ray, including multi-tenant cluster utilization
• Experience with fault tolerance and elastic training for long-running jobs - checkpointing strategy, straggler mitigation, preemption recovery
• Familiarity with autonomy or robotics data (ROS, OpenCV, multi-sensor log formats)
• Don’t meet every single requirement? If you’re excited about this role but your past experience doesn’t align perfectly with every qualification in the job description, we encourage you to apply anyway. You may be just the right candidate for this or other roles