Topics
former
AI
Amazon
Image Credits:Ariya Sontrapornpol / Getty Images
Apps
Biotech & Health
Climate
Image Credits:Ariya Sontrapornpol / Getty Images
Cloud Computing
Commerce
Crypto
endeavour
EVs
Fintech
fundraise
Gadgets
Gaming
Government & Policy
Hardware
Layoffs
Media & Entertainment
Meta
Microsoft
Privacy
Robotics
Security
societal
Space
Startups
TikTok
Transportation
Venture
More from TechCrunch
outcome
Startup Battlefield
StrictlyVC
newssheet
Podcasts
telecasting
Partner Content
TechCrunch Brand Studio
Crunchboard
get through Us
For many companies working on AI models with coating in the physical world , datum gift the big opportunity . It ’s also the biggest hurdle they face , as nicely labeled and sporty real - world data point is as promptly available as hen ’s tooth , and the costs and crusade required to gather and clean up data can be huge .
Bifrost , a 3D data - generation political program , believes its tech can help robotics and industrial company figure out at least one part of that problem : the time require to take aim AI models . The startup , establish out of San Francisco , says its platform lets companies give simulated 3D worlds to instruct their AI models and help their robots adjust to young objects , tasks , and surroundings within hour instead of months .
The company said on Wednesday that it has raised $ 8 million in a Series A funding one shot led by Carbide Ventures .
“ Most of our customers need vast sum of money of real - world data to train AI models , ” co - beginner and CEO Charles Wong said in an single interview with TechCrunch . “ This often mean they would have to deploy fleets of automaton across C of locations , collect millions of hours of footage , manually label the data , and implement rigorous quality checks to reduce human misplay and preconception . This approach is brutish . It costs millions , film years , and try out most impossible to scale . ”
Wong co - founded the firm with Aravind Kandiah in 2020 . Wong previously worked on AI percept models for self - tug cars at NuTonomy , an MIT spinout that worked on self - driving vehicles and autonomous mobile golem . Meanwhile , Kandiah previously build a aesculapian AI organisation that discover early signal of cecity and diabetic retinopathy .
“ It did n’t take long for us to realize something cardinal : AI and robotics need tremendous amounts of high - lineament data to function well . And that datum is essential , ” Kandiah told TechCrunch . “ It can make or break the execution and potential of these systems . So we join force to start Bifrost with a single finish : work the information problem , so AI and robotics can last tackle the complex challenge of the physical world . ”
Bifrost claims it is unlike from its competitors because its political platform does n’t take a squad versed in creating 3D pretence to generate such data . This , Wong say , gives AI organize a substantial advantage , allowing them to develop AI system for tasks like patrolling contested H2O with autonomous sauceboat without needing to hire a 3D squad .
Join us at TechCrunch Sessions: AI
Exhibit at TechCrunch Sessions: AI
“ Nvidia ’s Omniverse tools , by contrast , require a dedicated three-D team just to operate , ” Kandiah said , add up that Bifrost enable engineers in various heavy industries to learn AI systems young skills and accomplish more faster .
Bifrost ’s mathematical product is currently in a closed in beta with select heavy industry partners . The startup will apply the fresh immediate payment to fund the political platform ’s public launch in the coming months , as well as to lease more staff to accelerate up merchandise development .
Wong say that the company ’s basal market is the U.S. , but it is also realize impulse in Japan thanks to the body politic ’s significant industrial sphere . The startup engender revenue via an annual subscription model .
The political platform ’s primary drug user admit AI developer who specialize in make robotics systems , computer vision , and perception example for applications in industry such as robotics , aerospace , defense , maritime , geospatial , and industrial automation . Its objective customers are big industrial companies , government organisation , and startup in the growth to late stages , all of which would have teams focused on developing physical AI solutions in their several field .
“ We are initially rivet on military mission - vital , heavy industrial applications . By 2025 , we direct to inflate platform availableness . … Looking ahead , we plan to support a broad spectrum of commercial robotics use cases , especially as robotics applications have been quickly emerging across nearly every major sector and industriousness , ” Kandiah say .
The Series A brings Bifrost ’s full capital raised to $ 13.7 million . The round also saw involvement from Airbus Ventures , Peak XV ’s Surge , Wavemaker Partners , MD One , and Techstars . The outfit has 22 staff in the U.S. and Singapore .
