Topics
Latest
AI
Amazon
Image Credits:DataCrunch / CEO Ruben Bryon
Apps
Biotech & Health
Climate
Image Credits:DataCrunch / CEO Ruben Bryon
Cloud Computing
Commerce Department
Crypto
DataCrunch team in Finland.Image Credits:DataCrunch
endeavour
EVs
Fintech
Fundraising
Gadgets
back
Government & Policy
Hardware
layoff
Media & Entertainment
Meta
Microsoft
Privacy
Robotics
Security
Social
Space
startup
TikTok
Transportation
Venture
More from TechCrunch
Events
Startup Battlefield
StrictlyVC
Podcasts
Videos
Partner Content
TechCrunch Brand Studio
Crunchboard
Contact Us
A fledgling startup is setting out to become one of Europe ’s first “ AI compute ” hyperscalers , with renewable vigour work a pivotal part in its pitch to prospective customers .
The AI Au charge has goad unprecedented demand for “ compute , ” which refers to the processing major power , infrastructure , and resources call for for tasks such as running algorithmic rule , executing machine scholarship models , and processing data . One of the big donee of this demand has been Nvidia , emerging as a$3 trillion powerhouseoff the back of demand for its GPU ( graphics processing units ) and associated AI hardware .
In bicycle-built-for-two , an industry of swarm base provider has sprung up off the back of Nvidia , raising bucketful loads of Johnny Cash en road . In the U.S. , we ’ve seen thelikes of LambdaandCoreWeave hit lofty billion - dollar mark valuationsto expound their information center surgical operation . Now , Finnish startupDataCrunchis throw its hat into the ring , touting itself as one of the “ few serious players ” in the space with all operations in Europe .
“GPU-as-a-service”
Founded in 2020 by CEORuben Bryon , DataCrunch — like its match — sells GPUs “ as - a - armed service , ” promise to reduce the costs for AI processing . The troupe today enjoin it has raised $ 13 million in seed funding , constituting $ 7.6 million in fairness financing from backer such as byFounders , J12 Ventures , andAiven co - founderOskari Saarenmaa . The remaining $ 5.4 million debt section hails fromLokalTapiolaandNordea .
While it ’s slightly unusual for a seminal fluid - stage startup to elevate such a significant part as debt , DataCrunch has done this for the exact same reason that others in the space , such as CoreWeave , have also been raisinghefty amounts of debt . It ’s all about using physical assets — for instance Nvidia GPUs — as collateral to guarantee loans , rather than giving away more equity .
It ’s also more effective to fix big buckets of capital this elbow room , as the banking concern can but take away the GPUs if thing go abdomen - up for DataCrunch . For those who manipulate the handbag strings , it ’s much less high-risk than invest in a pure - fun SaaS startup , for instance .
“ Given the business that we ’re in , our main expenses for enlargement are capex [ working capital expenditure ] drive , ” Bryon told TechCrunch . “ This is the logical path to go about it , and as we grow , additional access to that financing becomes available . ”
Join us at TechCrunch Sessions: AI
Exhibit at TechCrunch Sessions: AI
This new round takes DataCrunch ’s entire funding raised since origination to $ 18 million , and will go some style toward help it build out its infrastructure to support Nvidia ’s late servers and clusters , includingthe sheeny new H200 GPU . In good turn , this will help it grow a customer base that not only includes collective client such as Sony , but item-by-item AI researchers working at the like of OpenAI .
“ That has always been an important food market for us , and I think that this ‘ individual ’ mart has been left behind by many , ” Bryon said . “ For me , in person , it ’s important — at the weekend , I ’m often using our own serving , and have been since the commencement . ”
Indeed , flexile , on - demand pricing is a far more alluring proposition for independent researchers and developers who might just need a fiddling bit of compute for personal or university projects .
“ multitude who are consider for a Masters or a Ph.D. — that ’s a segment we want to ride out connected to because it ’s often people who are a few years aside from doing something really with child , ” Bryon said .
sneak them in now , and draw the rewards later when they hit the big clock time . That ’s the cosmopolitan gist .
But there ’s no escaping the giant elephant in the room , one that all the cloud companies are cause to reckon with : thegargantuan amount of energyrequired topower this AI gyration .
Green machine
Part of DataCrunch ’s “ advantage ” is the fact that its data center are locate in the Finnish capital , Helsinki , and Iceland — a countryrunning on 100 % renewable energyfor yr already .
“ In Helsinki , we can support to dark-green energy from the power grid , ” Bryon say . “ And presently , in one of our two Finnish data centers , the waste warmth is captured to heat up Helsinki itself . In Iceland , we have the reward that the ambient air temperature is always low , while the energy admixture on the grid is already 100 % green . So Iceland is pretty much one of the light-green shoes in the world to have these form of operations . ”
This will be a big focal power point for the fellowship moving forth . While it plan to offer its overhaul to any caller globally , it will mostly remain anchored in the Nordics and Iceland . “ Perhaps in the future we ’ll look at Canada if we can find desirable locations , where we can have a standardised advantage in terms of carbon footprint of our operations , ” Bryon say .
It ’s these “ green ” credentials that DataCrunch desire will also congeal it asunder from other European rivals : company likeFlexAI in France , which recently exited stealthwith $ 30 million in seed funding ; andNebius , which recentlyemerged from the ashesof Russian internet heavyweight Yandex andhas just become a public companyagain .
There is a trade - off here , though : While down in the mouth latency is often one of the bounteous merchandising point for AI compute provider , DataCrunch is n’t necessarily go to be in that bucketful , which means it will be considerably befit for a particular form of work load .
“ Our scheme is such that we ’re not going to be the provider with the absolute lowest latency due to being in 100 location around the macrocosm , ” Bryon said . “ We are more focused on the compute that does n’t have that rigorous latency necessity . We can still have a nice enough latent period though ; it might not be 10 milliseconds , but it will still be something like 100 milliseconds . ”
It ’s also deserving remark that DataCrunch ’s data centers are in shared “ co - location ” facilities for now , but the fellowship says it ’s plan to start make out its own datum centers in 2025 — something it will call for significantly more capital for .
“ I want us to be on a way toward go public with this company , and we ’ll call for admittance to plenty more capital to keep expanding the fellowship , ” Bryon say .
