We are looking for exceptional talent to join our small but ambitious lab in Tokyo. Browse the open positions below and apply for the role that fits you best. エンタープライズ・ソリューションのポジションにご興味がある方は ...

A showcase of our post-training process, with web search and code execution built in.

Join the team behind Sakana Chat 🐟, Sakana Marlin 🐬, and Sakana Fugu 🐡 Build the operating backbone of Sakana's product portfolio. Tokyo (Hybrid) Full-time Japanese (Business) & English (Business) ...

管理画面のイメージ。*この画像は架空のデータを使ったイメージです。今後予告なく変更される可能性があります。 （English Announcement Below.） Sakana AIは本日、エンタープライズのお客様向け ...

2026年10月5日、Sakana AIは昨年に続き2回目となるエンジニア向けOpen Houseを開催しました。昨年はApplied Research Engineerに絞った内容でしたが、今年はApplied Research Engineer / Forward Deployed Engineer / Software Engineer / Solution Engineer / ...

開戦から約5ヶ月、ホルムズ海峡は「石油市場、史上最大の供給障害」の震源であり続けている。軍事力で圧倒しているはずの米国はなぜ出口を失ったのか。封鎖が長引く構造と今後の ...

Following our 2024 research on evolutionary model merging, a technique for “mixing to create” better models from existing ones (left), we are now tackling the challenge of “mixing to use” frontier ...

We are pleased to present our research at ICML 2026, “Bridging Spherical Black-Box Optimizers”. Full Paper: https://arxiv.org/abs/2606.25761 When optimizing ...

For over a decade, we’ve accepted that end-to-end backprop is the only way to train deep networks. But holding the entire network in memory all at once is why AI training is hitting a resource wall.

A longstanding goal of AI research has been the creation of AI that can learn indefinitely. One tantalizing path toward that goal is an AI that improves itself by rewriting its own code, including any ...

For the past 300,000 years, Earth has had only one form of advanced intelligence on it: humans. With the recent advent of AI foundation models, some believe we are at the dawn of a new kind of ...