Job Description
This machine learning systems researcher job at Lightmatter models how AI training and inference run on systems built with the company’s Passage silicon-photonics interconnect, designs rack-scale reference platforms and publishes ML systems research, based in Mountain View, California.
At a glance
- Position: Machine Learning Systems Researcher, Machine Learning team (full time)
- Where: Mountain View, California, USA (flexible, hybrid workplace)
- Pay: Base salary $186,000 to $240,000 a year, plus equity
- Apply: open until filled (re-checked 24 October 2026)
What the ML systems researcher works on
- Build tools that model the performance of Passage-enabled systems on training and inference for current AI models
- Design reference platforms for customers with the product and solutions-architecture teams
- Act as the team’s machine learning expert and keep up with ML research
- Publish papers at leading conferences and journals, and support white papers and technical blog posts
Why this role matters
Large-model training is now limited as much by moving data between chips as by compute. Lightmatter builds optical interconnects to ease that bottleneck, and someone has to show, with credible models, what a photonic fabric changes for real LLM workloads. This role does that and publishes the results.
What Lightmatter is looking for
- A PhD in computer science, electrical engineering or a related field, and 2–5 years of industry or research experience
- A solid background in ML systems research, performance evaluation and performance modelling
- Hands-on deep learning experience and extensive knowledge of LLM training and inference frameworks
- Knowledge of AI systems hardware (GPUs, TPUs) and networking (Ethernet, InfiniBand, NVLink), their topologies and trade-offs
- Academic publications, and the ability to own projects from start to finish
Nice to have
- First-principles problem solving in ambiguous settings
- Clear communication of complex results to partners and internal teams
Pay and location
The posting gives a base salary range of $186,000 to $240,000 a year, depending on location, experience and education, plus a new-hire equity grant and annual equity. Benefits include health cover, retirement matching and paid family leave. Candidates must be able to comply with US export control rules.
How to apply
Apply through the official Lightmatter job posting. Lightmatter’s posting does not give a closing date, so the role is open until filled; ResearchJobs.in will re-check this listing on 24 October 2026. Check the posting for location and work-authorisation details before you apply.
See all our artificial intelligence jobs, or browse more research jobs on ResearchJobs.in.
Hiring institution: Lightmatter
Official advertisement: job-boards.greenhouse.io
How to prepare for this application
- Prepare a performance model you built for distributed training or inference, and show where it matched and missed measurements
- Be ready to estimate, on a whiteboard, communication time for tensor, pipeline and expert parallelism on a given topology
- Read Lightmatter's public material on Passage and think about which parallelism strategies benefit most from higher bandwidth
- Know current LLM serving and training stacks (for example Megatron-LM, DeepSpeed, vLLM) well enough to discuss their bottlenecks
- Bring your publication list and one paper you can explain in depth
About Lightmatter
Lightmatter is a US company, founded in 2017 and based in Mountain View, California, that develops silicon-photonics technology for AI data centres, including the Passage photonic interconnect.