L

Lightmatter

Photonic computing hardware for AI acceleration

Principal Machine Learning Infrastructure Researcher

Full-TimeUpdated on 10/23/2026
$300k - $350k+ Equity grants
Expert
PhD
Mountain View, CA, USA
In PersonFlexible, hybrid workplace model is mentioned, but no office/home attendance schedule is specified.
AI & Machine LearningAI Research

About the job

Requirements
  • A PhD in Computer Science, Electrical Engineering, or a related field and at least 10 years of industry or research experience in the field.
  • A solid background in machine learning systems research and experience with performance evaluation and performance modeling.
  • Hands-on experience with machine learning or deep learning methods and extensive knowledge of large language model training and inference frameworks.
  • Academic publications in relevant journals and industry conferences.
  • Strong knowledge of AI systems, including compute technologies such as GPUs and TPUs and networking technologies such as Ethernet, InfiniBand, and NVLink, including their topologies, paradigms, and trade-offs.
  • A proven track record leading a team to deliver published results.
Responsibilities
  • Stay up to date on the latest advancements in machine learning research.
  • Build tools to model the performance of Passage-enabled systems running training and inference workloads for the latest AI models.
  • Collaborate with product and solutions architecture teams to design reference platforms for customers powered by Lightmatter technology.
  • Function as the machine learning expert on the team.
  • Publish academic papers at leading industry conferences and journals and support publication of whitepapers, blog posts, and other marketing collateral.
  • Lead a small team of machine learning engineers.
Desired Qualifications
  • Creatively solve problems and apply first-principles thinking to generate novel ideas, particularly in ambiguous, evolving environments.
  • Communicate complex research outcomes and technical concepts to partners and internal stakeholders.
  • Work across research, product, and business functions to bridge gaps and accelerate outcomes.
  • Adapt to a high-speed, high-precision environment.
  • Manage engineers and lead projects, supporting the team and maintaining high standards and collaboration.

About the company

Lightmatter develops photonic computing hardware to speed up AI workloads using light-based processing. Their photonic systems run AI tasks faster and use about 90% less energy than traditional processors. They target AI and autonomous vehicle workloads and collaborate with universities to advance electro-photonic systems. Revenue comes from hardware sales, software licenses, and partnerships, with a goal of bringing scalable photonic computing to market for faster, energy-efficient AI and AV applications.

Company Size

201-500

Company Stage

Series D

Total Funding

$822M

Headquarters

Mountain View, California

Founded

2017

Get referred to Lightmatter

See people who can refer or advise you

Simplify Jobs

Simplify's Take

What believers are saying

  • NVIDIA added Lightmatter to NVLink Fusion on May 21, 2026.
  • Open CPX MSA joined Lightmatter on September 17, 2026, legitimizing its architecture.
  • Lightmatter plans Guide DR sampling in Q4 2026 and CPX kits in Q1 2027.

What critics are saying

  • Passage L20 CPX evaluation kits ship in Q1 2027, delaying revenue conversion.
  • Broadcom, Marvell, and NVIDIA can bundle competing interconnects into incumbent platforms.
  • If hyperscalers standardize on electrical or pluggable optics, Lightmatter's CPO thesis collapses.

What makes Lightmatter unique

  • Lightmatter's Passage platform hit 1.6 Tbps per fiber in March 2026.
  • Guide DR reaches 51.2 Tbps per laser module with liquid cooling.
  • vClick Optics and bidirectional CPX cut fiber, connectors, and packaging complexity.

Help us improve and share your feedback! Did you find this helpful?

Benefits

  • Health Insurance
  • Dental Insurance
  • Vision Insurance
  • Life Insurance
  • Disability Insurance
  • Health Savings Account/Flexible Spending Account
  • Unlimited Paid Time Off
  • Flexible Work Hours
  • Remote Work Options
  • Paid Vacation
  • Paid Sick Leave
  • Paid Holidays

Growth & Insights and Company News

6 month growth

3%

1 year growth

0%

2 year growth

1%
PR Newswire
Oct 13th, 2026
Lightmatter unveils 204.8Tbps CPO switch tray with Flex and 102.4Tbps XPU tray at OCP

Lightmatter, a photonic computing company, unveiled two optical systems built to Open Compute Project (OCP) standards at the 2026 OCP Global Summit in San Jose. The first is a 204.8 Tbps co-packaged optics (CPO) switch tray in one rack unit, developed with Flex as a production validation system. It uses eight Passage CPO engines and is liquid-cooled. The second is a 102.4 Tbps XPU tray that holds four accelerators and two CPUs. Optical links replace copper, and Lightmatter says it is the first compute tray to carry four XPUs with that bandwidth in one rack unit. Lightmatter says its switch systems are two to four times more efficient in rack faceplate density than conventional designs. SENKO and Corning supplied fibre connectivity.

PR Newswire
Sep 17th, 2026
Lightmatter unveils first bidirectional CPX optical engine, cutting AI cluster fiber needs by 50%

Lightmatter has joined the Open CPX Multi-Source Agreement and launched the Passage L20 CPX, the industry's first Open CPX module for AI infrastructure with bidirectional optical link technology. The solution delivers 6.4 terabits per second in a single module whilst halving required fibre and connectors. The bidirectional technology combines transmit and receive on a single fibre by assigning each direction its own wavelength of light. In a 512-GPU scale-up pod, this saves more than 200 miles of fibre, reducing the total from approximately 130,000 fibres to 65,000, with around 16,000 fewer connectors. The reduction saves about 15 per cent of total scale-up interconnect network cost. Module evaluation kits are anticipated to ship in the first quarter of 2027.

Yahoo Finance
Jun 3rd, 2026
Lightmatter joins NVIDIA NVLink Fusion ecosystem with photonic interconnects for AI infrastructure

Lightmatter, a photonic computing specialist, has joined NVIDIA's NVLink Fusion ecosystem to deliver high-performance optical connectivity for AI infrastructure. The company will provide Co-Packaged Optics and Near-Packaged Optics products compatible with NVIDIA's optical and SerDes technologies. Lightmatter's bi-directional optical link architecture will create a unified platform for semi-custom AI factories whilst reducing fibre and connector requirements by 50%. The technology enables semi-custom XPUs to connect with NVIDIA switch silicon through Lightmatter's products, creating seamless high-bandwidth connectivity for chips from various suppliers. The partnership combines Lightmatter's Passage CPO solutions with NVIDIA's NVLink Fusion architecture. Analysts view this collaboration as a critical milestone in the maturation of co-packaged optics, providing hyperscalers with a blueprint to overcome traditional I/O bottlenecks and scale AI clusters.

Business Wire
May 21st, 2026
Lightmatter unveils Guide DR: liquid-cooled laser NIC quadruples rack density for AI datacentres

Lightmatter has unveiled Guide DR, a high-density laser network interface card that quadruples rack density compared to conventional external laser small form factor pluggables. The liquid-cooled LNIC relocates the light source from the faceplate into the chassis, solving scaling bottlenecks whilst leveraging existing cooling infrastructure. Built to OCP NIC 3.0 dimensions, a single Guide DR module can support up to 51.2 Tbps of aggregate bandwidth. Four modules in a 1 RU switch tray enable 204.8 Tbps of switching bandwidth, eliminating the need for larger 4 RU chassis with liquid-cooled lasers. The module supplies 200 mW of optical power per fibre across up to 64 fibres, driving 256 lanes at 200G each, with zero front-panel footprint. Guide DR will begin sampling in Q4 2026.

Associated Press
Apr 27th, 2026
Lightmatter hires ex-Google AI director Roy Kim to scale photonic interconnects globally

Lightmatter, a photonic computing company, has appointed Roy Kim as Vice President of Product to lead deployment of its photonic interconnect platform. Kim brings over 15 years of AI infrastructure experience from Google, AMD and NVIDIA. At Google since 2021, Kim served as Director of AI Infrastructure Product Management. He previously led Data Center GPU Product Management at AMD and spent eight years at NVIDIA in AI and high-performance computing roles. At Lightmatter, Kim will oversee product roadmap, management, marketing and strategy. The appointment follows significant product launches, including the Guide light engine in January and the Passage L20 optical engine in March. Lightmatter's technology aims to eliminate data bottlenecks in AI data centres through photonic interconnects.