MLOps Engineer, LLM Systems (Serving, GPU Kernels, Profiling)

This role is for one of our clients
Compensation: $90-$120 per hour
Join a leading AI lab's cutting-edge GenAI team and help build foundational AI models from the ground up. We're seeking MLOps Engineers with hands-on experience in large language model infrastructure across any of four areas: GPU kernel programming, performance profiling and trace analysis, debugging accelerated and distributed workloads, and high-throughput inference serving. This role involves AI model training and evaluation work, including writing and assessing MLOps and ML systems tasks and solutions to generate high-quality training data for frontier AI systems.
Requirements
Key Responsibilities

- Design challenging, domain-relevant tasks across four areas, GPU kernels, performance profiling, debugging, and inference serving, and write accurate, well-structured solutions to them.

- Guide research and engineering teams to close knowledge gaps and improve AI model performance on ML systems, training infrastructure, and framework-level topics.

- Evaluate MLOps and ML systems tasks and solutions, and provide clear, written technical feedback that stands up to reviewer scrutiny.

- Develop guidelines and detailed rubrics or evaluation frameworks covering kernel-level optimization, profiler output interpretation, distributed systems reasoning, and serving throughput and latency trade-offs.

- Collaborate with other subject matter experts to keep training data consistent and accurate.
Core Qualifications

- 2+ years of hands-on professional experience in ML systems, ML infrastructure, model serving, or GPU and accelerator performance engineering. This is a hands-on systems role rather than an applied modelling or data science one.

- Practical experience in at least one of the following, with more than one a strong plus: writing or optimizing custom GPU kernels (CUDA, Triton, Pallas); performance profiling and trace analysis (Kineto, torch.profiler, Nsight, XLA or JAX profiler); debugging distributed or accelerator-bound workloads; serving large language models at scale (vLLM, SGLang, TensorRT-LLM, Ray Serve, KV cache, paged attention, continuous batching).

- Working production experience with JAX and/or PyTorch. Framework-level depth is a strong plus: custom operators, distributed training (FSDP, DDP, DeepSpeed, Megatron), or compiler and graph-level work.

- Familiarity with modern accelerators such as A100, H100, B200 or TPU, and the ability to reason about throughput, latency and memory trade-offs.

- Demonstrable career progression.

- Ability to engage reliably for at least 40 hours/week during weekdays.

- Strong written communication skills and the ability to explain complex technical decisions clearly.