About Yodo Labs
We build spatial intelligence for the physical world, connecting images, video, 3D scans, and field records to real assets, locations, conditions, and risks. We've just closed our first round of funding and are building the team.
What you'll do
Own research directions on messy real-world data (camera, drone, LiDAR, point clouds, geospatial, inspection records): visual localization, asset matching, spatial registration, temporal re-identification, and anomaly detection. Take problems from paper to prototype to production — read the literature, test whether ideas actually work on our data, and turn the ones that do into systems that ship. You'll work directly with the founders, shape the technical roadmap, and publish strong results.
What we look for
- A strong foundation in CV, ML, 3D vision, or multimodal AI, with a track record of taking research from idea to working system (papers, products, or open source)
- Comfort with messy real-world data and the ability to drive work independently in an early-stage environment
- Working proficiency in either English or Japanese is enough; our team operates in both
Nice to have
SLAM, SfM, NeRF, 3D Gaussian Splatting, depth estimation, visual localization; vision-language models, segmentation, object detection; BIM/CIM, digital twins, asset management, infrastructure inspection; GPU / cloud training.
You don't need
A PhD, prior industry experience, or every skill above.
What you'll gain
Founding-team ownership of AI built for real physical environments, with access to industrial imagery, LiDAR data, and GPU resources — and the chance to see your research decisions play out in real systems, alongside the founders.
We sponsor Japanese work visas for international candidates and will support your relocation to Japan.
How to apply
Send a note to team@yodolabs.jp with the subject line "Research Scientist - [Your Name]." A short description of what you've worked on, along with a CV, GitHub, or paper if you have one, is all we need to start a conversation.