Senior Software Engineer - AI Inference

# Senior Software Engineer - AI Inference PerformanceNVIDIAUS, CA, Santa Clara Full-time $184,000 – $356,500 / yearAI tools:

Py Torch Tensor RT-LLMvLLMSGLangCUDACUTLASSTriton

Upload your resume once and Scout applies for you Apply with Scout Share this job:

Linked

InXSave image

This engineer will optimize large language and vision-language model inference on NVIDIA GPU systems. The work spans performance analysis, serving software, distributed runtimes, and GPU kernels, with results intended to improve production deployments and future platforms.
* Permanent* On-site
* More than 6 years
* Bachelor's or Master's Degree, or equivalent experience### Skills and expertise

PythonCUDAC++RustGPU performance profilingLLM/VLM inference optimization

Distributed systemsGPU architecture### Key responsibilities
* Analyze and optimize LLM and VLM inference workloads for latency, throughput, and efficiency.
* Profile workloads and improve host code, CUDA kernels, memory, communication, and scheduling.
* Build repeatable benchmarks and contribute performance improvements to inference engines.## How this job compares

Compared with 410 open senior machine learning jobs on October 10, 2026.Pay:
This job pays $184K to $357K a year. The 143 open senior machine learning jobs that publish US pay start at $186K on average, so this job starts in line with them.

Similar open jobs:
We Love AI Jobs lists 410 open senior machine learning jobs at 176 companies.

Remote work:
This job is on site. 90 of the 410 senior machine learning jobs (22%) are remote.

AI in this role:
High. AI is the core of the work: building, training or deploying AI systems. The role directly optimizes inference for LLMs and VLMs, including serving techniques and model execution. Hands-on work with AI inference performance is the core of the position.## Full description NVIDIA is the platform upon which every new AI-powered application is built. We are seeking a Senior Software Engineer – AI Inference Performance to advance innovative LLM and VLM inference.

You will push workloads toward practical performance limits on NVIDIA GPU-accelerated systems. Your work will span models, serving software, distributed runtimes, communication, CUDA kernels, and GPU architecture. Deliver measurable gains in latency, throughput, efficiency, and scale.

This is a hands-on role for an engineer who turns performance models and profiler data into working code. You will collaborate with model, framework, kernel, networking, and GPU architecture teams. You will contribute improvements to open-source inference engines and develop methods that others can reproduce. Your work will improve production deployments and help build future NVIDIA platforms.

What you'll be doing:
• Lead end-to-end analysis of LLM/VLM inference processes. Define representative prefill and decode workloads. Optimize time to first token, inter-token latency, P99 end-to-end latency, processing efficiency, and key-value (KV) cache capacity. For multimodal models, isolate preprocessing, encoder, and decoder costs.
• Build speed-of-light and roofline models to quantify performance headroom. Connect arithmetic intensity, bandwidth, occupancy, memory hierarchy, and communication costs to clear optimization hypotheses.
• Profile workloads using NVIDIA Nsight Systems, Nsight Compute, PyTorch Profiler, and custom instrumentation. Eliminate bottlenecks in host code, CUDA kernels, memory, communication, and scheduling.
• Tune serving hyperparameters and techniques such as batching, KV-cache management, quantization, speculative decoding, CUDA Graphs, and model parallelism. Choose them based on workload, hardware, model quality, and service-level objectives.
• Build and optimize performance-critical kernels, including attention,…