Staff Machine Learning Systems Researcher

Lightmatter•Mountain View, CA

23d•Hybrid

About The Position

Lightmatter is leading the revolution in AI data center infrastructure, enabling the next giant leaps in human progress. The company invented the world’s first 3D-stacked photonics engine, Passage™, capable of connecting thousands to millions of processors at the speed of light in extreme-scale data centers for the most advanced AI and HPC workloads. Lightmatter raised $400 million in its Series D round, reaching a valuation of $4.4 billion. We will continue to accelerate the development of data center photonics and grow every department at Lightmatter! If you're passionate about tackling complex challenges, making an impact, and being an expert in your craft, join our team of brilliant scientists, engineers, and accomplished industry leaders. Lightmatter is (re)inventing the future of computing with light! Lightmatter is seeking a Staff Machine Learning Systems Researcher to join our Machine Learning team. This team is focused on 1) inventing novel hardware systems using Lightmatter technology, and 2) modeling the performance of Lightmatter-enabled systems on both training and inference workloads. This is primarily a creative role where you need to think outside the box to push the boundaries of systems research with the power of silicon photonics at your fingertips. This role will work closely with our product, solutions architecture, and engineering teams to generate impactful performance analysis, create high-performance rack-scale designs, and publish cutting-edge ML systems research.

Requirements

PhD in Computer Science, Electrical Engineering, or a related field.
2 - 5 years of industry and/or research experience in the field.
Solid background in Machine Learning systems research and experience with performance evaluation and performance modeling.
Hands-on experience with machine learning / deep learning methods and extensive knowledge of Large Language Model training and inference frameworks.
Academic publications in relevant journals and industry conferences.
Must have knowledge of AI systems, including compute (GPUs, TPUs, etc.) and networking (Ethernet, Infiniband, NVLink, etc.) technologies, and their various topologies, paradigms, and trade-offs.
Ability to independently own projects from conception to completion.

Nice To Haves

Demonstrated ability to creatively problem-solve and apply first-principles thinking to generate novel ideas, particularly in ambiguous, evolving environments.
Exceptional skills in communicating complex research outcomes and technical concepts to partners and internal stakeholders, while embodying our value of speaking truth.
Ability to work across research, product, and business functions, rolling up your sleeves to bridge gaps and accelerate outcomes, while demonstrating the core value of "We are one Lightmatter."
Adaptable and versatile, thriving in a high-speed, high-precision environment.

Responsibilities

Stay up to date on the latest advancements in Machine Learning research.
Build tools to model the performance of Passage-enabled systems when running both training and inference workloads for the latest AI models.
Collaborate with our product and solutions architecture teams to design compelling reference platforms for our customers, powered by Lightmatter technology.
Function as the Machine Learning expert in the room.
Publish academic papers at top industry conferences and journals.
Help support the publication of whitepapers, blog posts, and other marketing collateral.