3 steps, 5 minutes, then you're done

Apply
Assessment
Recruiter Review

This remote job application is free, so give your best at every step of the process. Describe your real skills and experience, however small, because that's what makes a recruiter believe you're the right person. Take a few extra minutes, and give yourself a real chance.

Get remote job updates first Follow on LinkedIn
Home / AI Training Jobs / Jobs / MLOps Engineer Jobs

MLOps Engineer, LLM Performance (Remote) | $90–$120/hr | Contract

The remote MLOps Engineer job is a full-time hourly contract opportunity for ML systems and infrastructure engineers worldwide, offering $90–$120 per hour. The work involves developing and evaluating technical tasks covering accelerator kernels, performance profiling, distributed workload debugging, and large language model inference serving.

Quick Facts

  • Role: MLOps Engineer, LLM Performance (Kernels, Profiling, Serving)

  • Work Mode: Fully Remote

  • Contract Type: Hourly Contract

  • Pay: $90–$120 per hour

  • Openings: Not specified

  • Schedule and Hours: Full-time, 40 hours per week, weekdays; no conflicting engagements permitted

  • Location Restrictions: Worldwide; engagement arrangements depend on the applicant's country

Project Details

This project is intended for experienced ML systems engineers and MLOps professionals with practical knowledge of large language model infrastructure. The work applies engineering expertise in accelerator programming, kernel optimization, performance analysis, distributed computing, and high-throughput inference systems to the development and evaluation of AI training data.

The engagement is fully remote and requires 40 hours per week during weekdays, without conflicting professional engagements. Employment arrangements depend on the applicant's country and may involve an employer-of-record partner. Explore more remote WFA jobs for additional work-from-anywhere opportunities.

APPLY HERE

Job Information

  • Responsibilities: Create challenging ML systems tasks involving accelerator kernels, profiling, debugging, and inference serving. Develop accurate technical solutions, evaluate AI-generated engineering responses, provide expert feedback, and prepare detailed assessment guidelines covering optimization, distributed systems, and serving performance.

  • Qualifications: At least two years of professional experience in ML systems, ML infrastructure, model serving, or accelerator performance engineering. Applicants should demonstrate career progression, strong technical communication, and the ability to commit to 40 hours per week.

  • Technical Skills: Practical experience in at least one relevant area, including Pallas, Triton, CUDA, CUTLASS, JAX/XLA profiling, Kineto, torch.profiler, Nsight, distributed workload debugging, vLLM, SGLang, TensorRT-LLM, JetStream, KV cache, or paged attention. Production experience with JAX or PyTorch is also required. Familiarity with TPU, H100, B200, Trainium, or MI300 accelerators is relevant.

  • Language and Schedule: Strong written communication skills are required. The role requires reliable weekday availability for at least 40 hours per week.

  • Contract Type: Hourly contract with country-dependent engagement arrangements. Candidates must be available for a full-time commitment without conflicting engagements.

Compensation

The advertised compensation is $90–$120 per hour. The number of available openings is not specified.

The payment cycle and payment method are not specified in the listing. Applicants should confirm these details, along with the applicable engagement arrangement, during the application process.

Who Should Apply

This opportunity is suitable for ML infrastructure engineers, MLOps specialists, and systems engineers with hands-on experience building, optimizing, or operating large-scale machine learning infrastructure. Candidates should have practical expertise in at least one major area: accelerator kernel development, performance profiling, distributed workload debugging, or large language model serving.

Experience across multiple areas is a strong advantage. Additional familiarity with JAX, XLA, TPUs, custom operators, distributed training frameworks, compiler technologies, and accelerator performance trade-offs can strengthen an applicant's technical fit.

Professionals exploring AI Training Jobs can also review related opportunities on RemoteWFAJobs.

How to Apply

  1. Review the technical requirements, including the minimum professional experience and relevant ML systems specialization.

  2. Prepare details of practical projects involving kernel optimization, profiling, distributed computing, or LLM inference infrastructure.

  3. Submit an application through APPLY HERE and confirm the country-specific engagement arrangement, payment terms, and full-time availability requirements during the hiring process.

FAQ

Is this MLOps Engineer job really remote?

Yes. The position is fully remote and open to candidates worldwide, with engagement arrangements depending on the applicant's country.

Is previous ML systems experience required?

Yes. Applicants need at least two years of professional experience in ML systems, ML infrastructure, model serving, or accelerator performance engineering.

Is a degree necessary for this remote position?

Not specified. The listing focuses on professional experience, practical technical skills, and demonstrable career progression rather than stating a formal degree requirement.

How many openings are available?

Not specified. The listing does not provide a total number of available positions.

Closing

This remote engineering opportunity focuses on LLM infrastructure, accelerator performance, kernel development, and high-throughput inference systems. It offers $90–$120 per hour for experienced ML systems professionals who can commit to 40 hours per week.

For more opportunities, browse RemoteWFAJobs remote positions and explore the available AI training opportunities.

#RemoteWFAJobs #MLOpsJobs #LLMEngineering #RemoteJobs #WorkFromHome