Principal ML Performance Engineer – GPU Optimization

Proxima

. Profile and optimize training and inference for structural and generative models, including transformers, diffusion, and geometric deep learning .

Posted 9/23/2026full-timeRemote • New York • United StatesLeadWebsite

Core Competencies

Role fit

Core Competencies

Use this summary to align your resume positioning with the role.

Demonstrates expertise in optimizing machine learning systems and performance engineering, with a strong focus on distributed training and GPU cluster management. Proficient in developing benchmarks and profiling tools to enhance AI and data-generation platforms.

Highest-signal resume keywords

Machine Learning Systems OptimizationCUDA and Triton ProficiencyDistributed Training at Multi-Node ScaleDeep Knowledge of PyTorch InternalsPerformance Engineering

ATS Keywords

Tailor your resume

Applicant Tracking System Keywords

Tip: use these terms in your resume and cover letter to boost ATS matches.

Hard Skills

CUDATritonPythonC++Distributed TrainingProfiling and BenchmarkingMemory Bandwidth OptimizationMixed Precision TrainingTensor ParallelismPipeline Parallelism

Soft Skills

Technical Direction SettingMentoring EngineersInfluencing Research Teams

Tools & Technologies

GCPNsightTorch.compileTensorRTXLAFSDPDeepSpeed

Industry Keywords

High-Performance ComputingGenerative ModelsTransformersGeometric Deep LearningProximity TherapeuticsProtein-Interaction Discovery

Tech Stack

Tools & technologies

Google Cloud PlatformNode.jsPythonPyTorchC++

About the role

Key responsibilities & impact

Profile and optimize training and inference for structural and generative models, including transformers, diffusion, and geometric deep learning
Write and tune custom kernels using CUDA and Triton
Use compilers such as torch.compile, TensorRT, and XLA when beneficial
Scale distributed training across 32–64 nodes using FSDP, DeepSpeed, tensor parallelism, pipeline parallelism, and mixed precision
Reduce inference cost by optimizing memory scaling for large complexes, improving diffusion sampling efficiency, batching ragged inputs, and maximizing throughput across up to 1000 GPUs
Manage GPU cluster efficiency on GCP, focusing on scheduling, utilization, spot strategy, and cost reporting
Develop benchmarks and profiling tools for the research team
Help develop Proxima’s AI and data-generation platform for proximity therapeutics and protein-interaction discovery

Requirements

What you’ll need

Minimum of 6+ years experience in ML systems, HPC, or performance engineering
BS/MS/PhD in CS, EE, or related field
Ability to set technical direction beyond coding, select infrastructure, influence research teams, and mentor engineers
Deep knowledge of PyTorch internals with hands-on experience profiling and fixing real bottlenecks
Experience with CUDA and Triton
Skilled at reading Nsight output
Strong understanding of memory bandwidth and occupancy
Experience with distributed training at multi-node scale
Strong proficiency in Python and C++
Able to name a model they made materially faster and quantify the improvement