Applied ML Systems Scientist -Scale LLMs

Cerebras Systems builds the world’s largest AI chip, 56 times larger than GPUs. Our novel wafer-scale architecture provides the AI compute power of dozens of GPUs on a single chip, with the programming simplicity of a single device. This approach allows Cerebras to deliver industry-leading training and inference speeds and empowers machine learning users to effortlessly run large-scale ML applications, without the hassle of managing hundreds of GPUs or TPUs.

Cerebras’ current customers include top model labs, global enterprises, and cutting-edge AI-native startups.

OpenAI recently announced a multi-year partnership with Cerebras, to deploy 750 megawatts of scale, transforming key workloads with ultra high-speed inference.

Thanks to the groundbreaking wafer-scale architecture, Cerebras Inference offers the fastest Generative AI inference solution in the world, over 10 times faster than GPU-based hyperscale cloud inference services. This order of magnitude increase in speed is transforming the user experience of AI applications, unlocking real-time iteration and increasing intelligence via additional agentic computation.

As an Applied Machine Learning Research Scientistat Cerebras, you will play a key role in turning modern machine learning techniques into scalable, high-performance systems. This role sits at the intersection of modeling and systemsfocused not on publishing new algorithms, but on understanding how they work and making them run effectively r work will directlyimpacthow large language models (LLMs) are trained,optimized, and deployed on one of the most advanced AI platforms in the world.

You will work closely with researchers and senior engineers to implement and improve workflows for LLM pretraining, fine-tuning, and reinforcement learning-based post-training. This includes building training pipelines, debugging complex system behaviors, improving model quality, anditerating ondata and evaluation strategies. Your contributions will help translatecutting-edge

ML ideas into reliable, production-ready systems that solve real-world problems.

This role is ideal for candidates who enjoy hands-on engineering, want to build deep intuition for ML systems, and are excited about working on LLMs and reinforcement learning in practice,not just in theory.

Responsibilities

• Apply post-training techniques (e.g. RLVR, RLHF, GRPO etc.) techniques to improve model performance.
• Build and maintain evaluation pipelines to measure model performance across tasks and domains.
• Debug issues across the ML stack, including data pipelines, training jobs, model outputs and mixed or lower precision computation.
• Collaborate with researchers to translate ML ideas into efficient, scalable implementation.
• Design, implement, and scale ML pipelines across all stages of LLM development (pretraining, fine-tuning, alignment).
• Work with large datasets, including dataset generation, filtering, and synthetic data approaches.
• Optimize training and inference workflows for performance, efficiency, and reliability.
• Contribute high-quality, maintainable code to shared ML infrastructure.

Skills & Qualifications

• Bachelor’s or Master’s degree in Computer Science, Engineering, or a related field.
• 0 - 5 years of experience (including internships, research, or industry experience) working with machine learning systems; we are hiring multiple positions for various levels.
• Strong programming skills in Python.
• Experience with ML frameworks such as PyTorch.
• Solid understanding of machine learning fundamentals.
• Familiarity with deep learning architectures, particularly transformers.
• Ability to read and understand modern ML papers and implement key ideas.

Preferred Skills & Qualifications

• Experience working with large language models (training, fine-tuning, and evaluation).
• Familiarity with reinforcement learning concepts.
• Experience with distributed training frameworks (e.g., FSDP, Megatron).
• Experience working with large-scale datasets and data pipelines.
• Experience debugging or optimizing ML systems for performance.
  
  • Contributions to meaningful codebases, projects, or open-source systems

People who are serious about software make…