AI Performance Engineer

Job not on LinkedIn

🔥 1 minute ago

🏄 California – Remote

🏄 California – Remote

⏰ Full Time

🟢 Junior

🟡 Mid-level

🤖 Artificial Intelligence

🦅 H1B Visa Sponsor

This company has sponsored H1B visas in the past.

👻 Ghost score 12%

What it means: how likely this posting is sitting open without active hiring.How to read it: 60% or higher means we think this is a ghost. Lower scores are less of a worry.How we score it: age, reposts, open roles vs team size, how fast jobs close, and whether pay is listed.Fine print: an estimate from public data, not a judgment of the company. Real jobs can score high, so treat it as a hint, not a fact.

Apply Now

Find Similar Remote Jobs

Set Up Email Alert For New Jobs

📊 Check your resume score for this job

Improve your chances of getting an interview by checking your resume score before you apply.

Upload Resume

Cornelis Networks

WebsiteLinkedInAll Job Openings

51 - 200 employees

🤖 Artificial Intelligence

🔧 Hardware

🏢 Enterprise

💰 $29M Series B on 2022-11

Artificial Intelligence • Hardware • Enterprise

Cornelis Networks is a leading provider of intelligent, scalable, high-performance interconnects designed for AI applications. The company specializes in delivering end-to-end purpose-built high-performance fabrics to commercial, scientific, academic, and government organizations. Cornelis Networks’ solutions are aimed at enhancing performance, scalability, and efficiency across hyperscale, cloud AI, and on-premises AI/HPC environments. Their offerings are known for scalable architecture, high bandwidth solutions, and universal compatibility with accelerators and GPUs. Originating as an Intel spin-off, the company is positioned to challenge existing technologies like InfiniBand and Ethernet, providing advanced interconnects that power modern AI infrastructure.

📋 Description

• Measure and optimize end-to-end performance for distributed AI workloads, including training and multi-node inference, across CPU/GPU clusters and Omni-Path, Ethernet, and InfiniBand fabrics • Benchmark, characterize, and tune open-source and industry workloads and benchmarks such as Llama and Qwen on compute, storage, and network hardware • Work with frameworks including Megatron-LM, TorchTitan, Composer, vLLM, and llama.cpp • Design and optimize distributed serving topologies, including sharded/replicated, tensor/pipeline parallel, and MoE expert placement • Optimize continuous/adaptive batching, KV-cache sharding/offload, and prefix caching • Validate RDMA/GPUDirect RDMA, congestion control, and collective/point-to-point tradeoffs during inference • Build proof points comparing Cornelis Omni-Path with competing interconnects • Translate performance data into narratives for sales, marketing, and customers • Instrument and visualize performance using Nsight Systems, VTune, perf, application timers, Prometheus, and Grafana • Evangelize best practices through briefs, READMEs, blogs, articles, and peer-reviewed publications • Partner with architecture, firmware, software, and lighthouse customers to turn lab results into field-proven wins

🎯 Requirements

• BS in CS/EE/CE/Math or related (MS preferred) • 2–3+ years running AI/ML at cluster scale • Proven ability to set up, run, and analyze AI benchmarks • Deep intuition for message passing, collectives, scaling efficiency, and bottleneck hunting for training and inference • Hands-on distributed training beyond single-GPU (DP/TP/PP, ZeRO, FSDP, sharded optimizers) • Experience with distributed inference architectures (replicated vs sharded, tensor/KV parallel, MoE) • Practical experience with PyTorch, DeepSpeed, Megatron-LM, RCCL/NCCL, vLLM, and llama.cpp • Comfortable with GCC/LLVM/Intel/OneAPI compilers and MPI stacks • Python and shell proficiency • Familiarity with Omni-Path/OPA, InfiniBand, Ethernet/RDMA/RoCE, and Linux performance tuning • Knowledge of NIC offloads, CQ moderation, pacing, and ECN • Excellent written and verbal communication • Preferred: Cluster scheduler experience (SLURM, PBS) • Preferred: GPU/communications profiling and tracing with Nsight Systems, Nsight Compute, ROCm tools/rocprof/roctracer, VTune, or perf • Preferred: HPC performance engineering or storage background (BeeGFS, Lustre, NVMe-oF)

🏖️ Benefits

• Full-Time (U.S. Benefit Eligible) • Fully supported remote work for employees residing within the United States • Travel to partner and client meetings, conferences, and company in-person collaboration and meetings (less than 10%)