Towaki Takikawa
Co-Founder CEO at Outerport
San Francisco, California, United States
Summary
🤩
Rockstar🎓
Top SchoolTowaki Takikawa is a founder and CEO with 11 years of experience building applied AI systems that bridge research and manufacturing, currently scaling Outerport to accelerate physical product development from concept to production. Previously a research scientist at NVIDIA, he focused on geometry processing and neural fields for 3D reconstruction and simulation, publishing 10+ papers (CVPR, ICCV, SIGGRAPH) with 4,500+ citations and contributing to influential open-source 3D ML tooling that has earned 8,000+ stars. He combines deep technical expertise—implementing core ray-tracing and neural field components in projects like NVIDIA Kaolin Wisp and supporting ICCV work such as GSCNN—with operational leadership honed by managing large multidisciplinary teams for autonomous vehicle projects. Based in San Francisco and a Y Combinator S24 founder, he works with Fortune 500 OEMs to deploy AI agents and orchestration tools that make technical documents and simulations actionable for engineers and manufacturers.
11 years of coding experience
7 years of employment as a software developer
Non-Degree Student Undeclared, Non-Degree Student Undeclared at Oregon State UniversityHigh School, High School at Crescent Valley High SchoolDoctor of Philosophy - PhD Computer Science, Doctor of Philosophy - PhD Computer Science at University of TorontoY CombinatorBachelor of Science Honours Computer Science, Bachelor of Science Honours Computer Science at University of Waterloo
English, Japanese
Github Skills (12)
mask-rcnn10
faster-rcnn10
pytorch10
computer-vision10
serf10
deep-learning10
sdf10
3d10
cuda9
documentation8
website-admin7
manage7
Programming languages (7)
Github contributions (5)
nv-tlabs/GSCNN
Jul 2019 - Nov 2019
Gated-Shape CNN for Semantic Segmentation (ICCV 2019)
Role in this project:
ML Engineer
ML EngineerContributions:10 commits, 2 PRs, 16 pushes in 4 months
Contributions summary:Towaki contributed to the documentation and website of the Gated-Shape CNN project. They updated the paper links, announced the paper's acceptance to ICCV 2019, and released the code. The user also removed an unused variable in the GSCNN network code, demonstrating their involvement in the project's development. These actions suggest a role in maintaining and promoting the project.
pytorchiccviccv-2019semantic-segmentationdeep-learning
NVIDIAGameWorks/kaolin-wisp
Jul 2022 - Jan 2023
NVIDIA Kaolin Wisp is a PyTorch library powered by NVIDIA Kaolin Core to work with neural fields (including NeRFs, NGLOD, instant-ngp and VQAD).
Role in this project:ML Engineer
ML EngineerContributions:30 commits, 44 PRs, 40 pushes in 5 months
Contributions summary:Towaki primarily contributed to the NVIDIA Kaolin Wisp library, focusing on core ray tracing functionality and neural field implementations. They added and modified channel interfaces within tracers, improving the rendering pipeline. Additionally, the user addressed bugs and implemented optimizations, specifically related to the loading and utilization of pre-trained models within the SDF framework.
pytorchnvidianeural-fieldsfieldswisp
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.