← Back to jobsJob ID col-cmwpbnh6a09mod8ha9ol9s0s8

ML Infrastructure Expert - Fully RemotePosted 1 day ago

ML Infrastructure Expert - Fully Remote | Up to $120/hrNew

Contract
Remote
2+ years
Competitive
Posted 1 day ago

Mercor is recruiting a contract MLOps Engineer for work focused on LLM performance. The role involves designing and evaluating technical tasks, developing expert guidelines and rubrics, and helping improve AI model performance across ML systems, compilers, and training infrastructure.

Posted
1 day ago
Source
ziprecruiter.com
Sector
Information Technology & AI

Opens the employer's application page in a new tab

Responsibilities

  • Design challenging tasks across kernels, profiling, debugging, and inference serving. Write accurate, well-structured solutions to expose where frontier models fail on accelerator code
  • Guide research and engineering teams to close knowledge gaps and improve AI model performance on ML systems, compilers, and training infrastructure topics
  • Evaluate ML systems tasks and solutions. Provide clear written technical feedback that stands up to expert review
  • Develop guidelines and detailed rubrics covering kernel 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

Qualifications

  • 2+ years of hands-on professional experience in ML systems, ML infrastructure, model serving, or accelerator performance engineering
  • Practical experience in writing or optimizing custom kernels (Pallas, Triton, CUDA, CUTLASS), performance profiling and trace analysis (JAX/XLA profiler, Kineto, torch.profiler, Nsight), debugging distributed or accelerator-bound workloads, serving large language models at scale (vLLM, SGLang, TensorRT-LLM, JetStream, KV cache, paged attention, continuous batching)
  • Working production experience with JAX and/or PyTorch
  • Experience with modern accelerators such as TPU (v4 to v6e), H100, B200, Trainium or MI300, and the ability to reason about throughput, latency, and memory trade-offs
  • Strong written communication skills and the ability to explain complex technical decisions clearly

Preferred

  • Framework-level depth: custom operators, sharding and distributed training (pjit/shard_map, FSDP, Megatron, MaxText), or compiler and graph-level work (XLA/StableHLO, torch.compile, MLIR)
  • torch.compile

Required skills

Accelerator Hardware B200Accelerator Hardware H100Accelerator Hardware MI300Accelerator Hardware TPUAccelerator Hardware TrainiumAccelerator Performance EngineeringCustom Kernel Optimization CUDACustom Kernel Optimization CUTLASSCustom Kernel Optimization PallasCustom Kernel Optimization TritonDistributed Training FSDPDistributed Training MaxTextDistributed Training MegatronDistributed Training pjit/shard_map

Full job description

Mercor is recruiting a contract MLOps Engineer for work focused on LLM performance. The role involves designing and evaluating technical tasks, developing expert guidelines and rubrics, and helping improve AI model performance across ML systems, compilers, and training infrastructure.

Responsibilities

  • Design challenging tasks across kernels, profiling, debugging, and inference serving. Write accurate, well-structured solutions to expose where frontier models fail on accelerator code
  • Guide research and engineering teams to close knowledge gaps and improve AI model performance on ML systems, compilers, and training infrastructure topics
  • Evaluate ML systems tasks and solutions. Provide clear written technical feedback that stands up to expert review
  • Develop guidelines and detailed rubrics covering kernel 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

Preferred

  • Framework-level depth: custom operators, sharding and distributed training (pjit/shard_map, FSDP, Megatron, MaxText), or compiler and graph-level work (XLA/StableHLO, torch.compile, MLIR)
  • Custom Operator Development
  • Sharding
  • Distributed Training
  • pjit
  • shard_map
  • FSDP
  • Megatron
  • MaxText
  • Compiler Development
  • Graph-Level Optimization
  • XLA
  • StableHLO
  • torch.compile
  • MLIR

Benefits

  • Remote work

Company

  • Expert staffing and frontier model training for enterprises
  • San Francisco, California, USA
  • https://www.mercor.com/
  • Founded in 2023
  • 501-1000 employees

Qualifications

  • 2+ years of hands-on professional experience in ML systems, ML infrastructure, model serving, or accelerator performance engineering
  • Practical experience in writing or optimizing custom kernels (Pallas, Triton, CUDA, CUTLASS), performance profiling and trace analysis (JAX/XLA profiler, Kineto, torch.profiler, Nsight), debugging distributed or accelerator-bound workloads, serving large language models at scale (vLLM, SGLang, TensorRT-LLM, JetStream, KV cache, paged attention, continuous batching)
  • Working production experience with JAX and/or PyTorch
  • Experience with modern accelerators such as TPU (v4 to v6e), H100, B200, Trainium or MI300, and the ability to reason about throughput, latency, and memory trade-offs
  • Demonstrable career progression
  • Strong written communication skills and the ability to explain complex technical decisions clearly
  • Framework-level depth: custom operators, sharding and distributed training (pjit/shard_map, FSDP, Megatron, MaxText), or compiler and graph-level work (XLA/StableHLO, torch.compile, MLIR)
  • 2+ years of hands-on professional experience in ML systems, ML infrastructure, model serving, or accelerator performance engineering.

Required skills

  • Accelerator Hardware B200, Accelerator Hardware H100, Accelerator Hardware MI300, Accelerator Hardware TPU, Accelerator Hardware Trainium, Accelerator Performance Engineering, Compiler and Graph-Level Optimization MLIR, Compiler and Graph-Level Optimization XLA/StableHLO, Compiler and Graph-Level Optimization torch.compile, Custom Kernel Optimization CUDA, Custom Kernel Optimization CUTLASS, Custom Kernel Optimization Pallas, Custom Kernel Optimization Triton, Distributed Training FSDP, Distributed Training MaxText, Distributed Training Megatron, Distributed Training pjit/shard_map, Distributed and Accelerator-Bound Workload Debugging, JAX, ML Systems and ML Infrastructure, Model Serving for Large Language Models JetStream, Model Serving for Large Language Models SGLang, Model Serving for Large Language Models TensorRT-LLM, Model Serving for Large Language Models vLLM, Performance Profiling and Trace Analysis JAX/XLA Profiler, Performance Profiling and Trace Analysis Kineto, Performance Profiling and Trace Analysis Nsight, Performance Profiling and Trace Analysis torch.profiler, PyTorch

Experience

  • 2+ years

Job details

Employment type

Contract

Seniority level

2+ years

Date posted

Nov 20, 2026

Work mode

Remote

Job source

ziprecruiter.com

Company

ML Infrastructure Expert - Fully Remote

Information Technology & AI

Expert staffing and frontier model training for enterprises

  • San Francisco, California, USA
  • Information Technology & AI
  • 501-1000 employees
  • Founded 2023
  • mercor.com

Similar jobs

Matched on role, skills and location · live jobs only

See more Machine Learning, Infrastructure jobs

Tilt

1d ago

New

Data Science Manager, LATAM

Full-timeRemoteUnited StatesCompetitive

Tilt is hiring a Data Science Manager to lead data science across its Mexico and Brazil markets. The role combines hands-on model building with team leadership, setting strategy and standards, and partnering with credit leadership to connect modelling work to…

Why similarSame role

Cardinal Health

1d ago

New

AI/ML Engineer

Full-timeOn-siteUnited States$96K – $124K/yr

Cardinal Health is seeking a product-oriented AI/ML Engineer to design and deliver enterprise-scale Generative AI solutions.

Why similarSimilar titleMachine Learning & AI

just now

1d ago

New

Senior Data Scientist - Preferred Health Information Exchange - Fully REMOTE

TemporaryRemoteUnited StatesCompetitive

NC HIEA operates NC HealthConnex, North Carolina’s state-designated health information exchange. The Junior Data Scientist supports operational, analytical, and quality improvement initiatives by preparing and monitoring data workflows, developing dashboards…

Why similarSimilar title

Get new Machine Learning, Infrastructure jobs by email

Create job alert

ML Infrastructure Expert - Fully Remote | Up to $120/hr

ML Infrastructure Expert - Fully Remote · Competitive

Apply on company site