MLOps Engineer, LLM Systems · Mercor Mercor · 90-120/hr · remote in US, UK, CA · 2w ago 90-120/hr re

Key Responsibilities: Design and solve challenging, domain-relevant MLOps tasks across GPU kernels, performance profiling, debugging, and inference serving to generate high-quality AI training data. Guide research and engineering teams to close knowledge gaps and improve AI model performance on ML systems and training infrastructure. Evaluate MLOps tasks and solutions, providing clear, rigorous written technical feedback. Develop detailed rubrics and evaluation frameworks covering kernel optimization, profiler output interpretation, distributed systems reasoning, and serving throughput/latency trade-offs. Collaborate with subject matter experts to ensure consistency and accuracy across training datasets.

Core Qualifications: 2+ yearsof hands-on professional experience in ML systems, ML infrastructure, model serving, or GPU/accelerator performance engineering. Practical experience in at least one of: custom GPU kernel development (CUDA, Triton, Pallas); performance profiling and trace analysis (Kineto, torch.profiler, Nsight, XLA/JAX profiler); debugging distributed or accelerator-bound workloads; or large-scale LLM serving (vLLM, SGLang, TensorRT-LLM, Ray Serve, KV cache, paged attention, continuous batching). Production experience withJAX and/or PyTorch; framework-level depth (custom operators, FSDP, DDP, Deep Speed, Megatron) is a strong plus. Familiarity with modern accelerators (A100, H100, B200, TPU) and the ability to reason about throughput, latency, and memory trade-offs. Strong written communication skills with the ability to explain complex technical decisions clearly. Availability for a dedicated 40 hours/weekon weekdays with no conflicting engagements. Cincinnatus LLC is an Equal Employment Opportunity employer. All qualified applicants will receive consideration regardless of race, religion, gender, age, disability, or any other protected characteristic.