LLM Serving Engine Engineer

Description

Preferred Networks (株式会社Preferred Networks) — Japan's leading AI unicorn ($2B+ valuation, $308M raised), builder of the PLaMo family of Japanese-language LLMs and the energy-efficient MN-Core AI processor line (MN-3 ranked #1 on Green500 three times), with deep partnerships with Toyota, FANUC, and NTT — is hiring an LLM Serving Engine Engineer to build the software infrastructure that serves PLaMo LLMs using its upcoming MN-Core L1000 inference accelerator, based onsite in Tokyo, Japan. Preferred Networks will provide full visa sponsorship and relocation support for international candidates. This role sits at the intersection of AI serving infrastructure, hardware-software co-design, and LLM deployment at a company that controls its own silicon — a rare and technically extraordinary environment. You'll build and optimize the serving infrastructure for PLaMo models on MN-Core hardware, working closely with the chip and model research teams. Preferred Networks has pioneered energy-efficient AI compute since 2018 and is one of the few AI research companies in the world with vertically integrated hardware-to-model capabilities.

Required Skills