Staff Inference ML Runtime Engineer

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. About The Role The Inference ML Engineering team at Cerebras Systems is dedicated to enabling our fast generative inference solution through simple APIs powered by a distributed runtime that runs on large clusters of our own hardware. Our mission is to empower enterprises, developers, and researchers to unlock the full potential of our platform, leveraging its performance, scalability, and flexibility. The team works closely with cross-functional groups, including compiler developers, cluster orchestrators, ML scientists, cloud architects, and product teams, to deliver high-impact solutions that redefine the boundaries of ML performance and usability. As a Senior Software Engineer on the Inference ML Engineering team, you will play a key role in designing and implementing APIs, ML features, and tools that enable running state-of-the-art generative AI models on our custom hardware. You will architect solutions that enable seamless model translation and execution, ensuring high throughput and low latency, while maintaining ease of use. Your responsibilities will include leading technical initiatives, collaborating with other engineering teams to enhance the developer experience, enabling key ML features at scale, maintaining our speed advantage, achieving high throughput, and supporting a wide range of ML workloads. This role offers an opportunity to shape the evolution of our ML ecosystem while tackling complex technical challenges at the intersection of machine learning, software, and hardware. Responsibilities Drive and provide technical guidance to a team of software engineers working on complex machine learning integration projects. Design and implement ML features (e.g., structured outputs, biased sampling, predicted outputs) that improve performance of generative AI models at inference time. Design and implement high-throughput, low-latency multimodal inference models that support delivery of image, audio, and video inputs and outputs. Maintain our scalable serving backend for handling many concurrent requests per minute. Scale our inference service by implementing detailed observability throughout the entire stack. Analyze and improve latency, throughput, memory usage, and compute efficiency on the service and the implementation of various features. Optimize software to accelerate generative LLM inference by achieving high throughput and low latency. Stay up-to-date with advancements in machine learning and deep learning, and apply state-of-the-art techniques to enhance our solutions. Evaluate trade-offs between different approaches, clearly articulate design choices, and develop detailed proposals for implementing new features. Uncover, scope, and prioritize significant areas of technical debt across the software stack to ensure continued high quality of the inference service. Build and maintain robust a