MLOps Engineer, LLM Performance (Kernels, Profiling, Serving)
Posted 2 days ago
You'll complete your application on Mercor's career site.
About the Role
This position invites you to become part of a top-tier artificial intelligence research organization, contributing to the core infrastructure that powers state-of-the-art generative language models. Your work centers on advanced machine learning systems engineering and MLOps, where you will focus on scaling inference, optimizing performance, and managing distributed computing tasks. You will push the limits of modern compiler and accelerator technologies by designing, evaluating, and improving intricate technical challenges. The role is hands-on and requires expertise in systems debugging, performance analysis, and kernel-level development. By producing superior training data, you will help advance model evaluation and creation. You will act as a link between academic research and production deployment, helping to ensure the robustness and functionality of sophisticated AI architectures alongside a team of industry experts.
Requirements
- You must have a minimum of two years of relevant work experience in performance engineering, ML infrastructure, or systems engineering, with a primary focus on systems architecture rather than general applied data science.
- You are expected to be skilled in at least one of these domains: large-scale LLM inference management, debugging distributed workloads, performing trace analysis and performance profiling, or developing custom Triton or CUDA kernels.
- You must have practical, professional experience using PyTorch or JAX.
- It is advantageous to have experience with compiler-level development, distributed training techniques, or the creation of custom operators.
- You should possess extensive familiarity with high-end GPUs or TPUs and be able to evaluate the balance between memory usage, latency, and throughput.
- You need to be able to clearly document and communicate technical decisions in writing.
- You must be prepared to work a full-time, 40-hour weekly schedule.
Responsibilities
- You will develop complex technical assignments involving inference, profiling, and kernel programming, specifically creating adversarial tasks to identify vulnerabilities in accelerator code.
- You will analyze proposed technical answers and offer precise, expert-level critiques.
- You will instruct engineering and research staff on infrastructure and compiler-related matters while establishing detailed criteria for evaluating system performance and kernel efficiency.
- You will work with other specialists to verify the precision and uniformity of training data, helping to refine the underlying AI training architecture.
Benefits & Perks
- This is a 40-hour per week, full-time position providing the chance to collaborate with a prominent AI laboratory as an extended team member.
- The role is entirely remote and available to individuals worldwide.
- Engagement terms are handled through local partners to provide necessary support and maintain regulatory compliance in your country of residence.
Note: This listing is summarized and paraphrased by WorkNow from a publicly posted role. For the original wording and full details, please refer to the employer's official posting.
Similar Jobs
Senior Shopify Plus Developer
Hunt St
We are looking for a Senior Shopify Plus Developer to work remotely from the Philippines. The position centers on advanced storefront development, theme tailoring, and sophisticated integrations, with a commitment to maintaining rigorous technical excellence and meeting project requirements. You will handle diverse client demands, ranging from ongoing operational maintenance and retainer engagements to comprehensive project deployments. You will serve as a primary technical authority, guiding projects through the full lifecycle from early estimation to final launch validation. Your role involves generating reliable projections, addressing technical issues, and ensuring tasks align with established budgets and scopes. Success requires consistent communication with internal teams and leads to manage project hurdles and risks. This role suits an expert developer comfortable in a decentralized agency setting. You are encouraged to use advanced AI development tools as part of our forward-thinking strategy and contribute to internal R&D initiatives. The position blends deep technical execution with project oversight, demanding high-quality coding standards and transparent collaboration with clients and management.
Freelance Website Developer, Graphic Designer
Pramira
We are looking for a multi-talented freelance Web Developer and Graphic Designer to work remotely. This position suits a creative expert who can integrate technical web work with superior visual design. Your main tasks include managing our WordPress presence to keep the site fast, secure, and attractive, while also creating various marketing materials. You will assist multiple departments by delivering custom design solutions that match our standards in the telecommunications and engineering sectors. This role provides great flexibility, as you can organize your own hours within a freelance structure. You will partner with our team to run marketing initiatives, keep branding consistent across all channels, and complete various digital and print production assignments. We need someone who can work autonomously, communicate effectively, and document their hours, making this a great fit for a self-driven person interested in varied project work.
Senior Art Director
Construct Digital
This agency focuses on building data-backed brand stories that produce tangible commercial success. Through a mix of artistic flair and sophisticated technical solutions, the firm drives growth in sales, lead acquisition, and ROI. The service offering includes data analysis, UX design, content marketing, and modern AI integration to optimize engagement and performance. As a Senior Art Director, you will work alongside creative technologists, planners, and strategists to deliver impactful solutions for clients throughout the Asia-Pacific territory. This pivotal position requires a blend of aesthetic leadership, hands-on production, and strategic guidance within an environment that promotes experimentation and continuous workflow improvement.