Quick! Picture physical AI.
*waiting*
If you pictured a shiny humanoid with a face, blame Hollywood.
Meanwhile, plenty of machines that look nothing like us are already on the job.
Read about physical AI already at work & what it takes to make it last 👉 okt.to/9uN0jT
*waiting*
If you pictured a shiny humanoid with a face, blame Hollywood.
Meanwhile, plenty of machines that look nothing like us are already on the job.
Read about physical AI already at work & what it takes to make it last 👉 okt.to/9uN0jT
0
0
0
42
GTC Berlin is a wrap! ⚡
Our team is heading home with a lot of good conversations about AI token spend and local-first inference. Thank you to everyone who spent part of their two days with us at the hospitality suite.
Our team is heading home with a lot of good conversations about AI token spend and local-first inference. Thank you to everyone who spent part of their two days with us at the hospitality suite.
0
0
0
61
We’re hiring a Senior Software Engineer to own AI inference at scale!
If you enjoy a hard performance problem and have run models in production with engines like vLLM, we’d love to hear from you.
Learn more about the role and apply here 👉 okt.to/Ol4iod
If you enjoy a hard performance problem and have run models in production with engines like vLLM, we’d love to hear from you.
Learn more about the role and apply here 👉 okt.to/Ol4iod
0
0
0
95
That’s a wrap on day one at GTC Berlin!
Thanks to everyone who came by our hospitality suite to talk token costs and see the demos (and shout-out to our new DJs 🎧).
Now we’re off to the Van Loon with @weka, then back at the suite tomorrow from 8am 👉 okt.to/li5CSk
Thanks to everyone who came by our hospitality suite to talk token costs and see the demos (and shout-out to our new DJs 🎧).
Now we’re off to the Van Loon with @weka, then back at the suite tomorrow from 8am 👉 okt.to/li5CSk
0
0
0
41
As an NVIDIA DSX ecosystem partner, we integrated open-source NVIDIA Run:ai into PaletteAI.
Together, we help customers put more of their GPU capacity to work without taking on greater operational complexity.
How it works and how to try it 👉 okt.to/oMcNAZ
Together, we help customers put more of their GPU capacity to work without taking on greater operational complexity.
How it works and how to try it 👉 okt.to/oMcNAZ
0
0
1
212
We’re proud to be an NVIDIA AI Cloud Ready validated partner, and to give neoclouds, sovereign AI clouds, and AI-as-a-service businesses the reliability, consistency, and scale they need to succeed in the long run.
More about the validation 👉 okt.to/L6xV4w
More about the validation 👉 okt.to/L6xV4w
0
0
1
60
GTC Berlin is right around the corner, and we'll be waiting for you at Gibson Bar!
We’re bringing it all: a smart crew, insightful demos, yummy canapés and a live DJ experience. And on the 21st, you can join us and @weka aboard the Van Loon.
Our plans 👉 okt.to/rAbq0N
We’re bringing it all: a smart crew, insightful demos, yummy canapés and a live DJ experience. And on the 21st, you can join us and @weka aboard the Van Loon.
Our plans 👉 okt.to/rAbq0N
0
0
0
55
Hybrid AI is going to keep evolving, but three themes will persist: placement, economics and control.
And they can’t be separated. The lowest-cost endpoint is irrelevant if it can’t meet the quality threshold.
Our CEO Tenry Fu goes deeper 👉 okt.to/IlYKqr
And they can’t be separated. The lowest-cost endpoint is irrelevant if it can’t meet the quality threshold.
Our CEO Tenry Fu goes deeper 👉 okt.to/IlYKqr
0
0
0
35
Getting AI factories live is only the start. Running them for years takes reliability, consistency, and scale, which is the foundation PaletteAI provides.
We're proud to be an NVIDIA AI Cloud Ready Validated Partner, and we talk more about it here 👉 okt.to/rJxaRE
We're proud to be an NVIDIA AI Cloud Ready Validated Partner, and we talk more about it here 👉 okt.to/rJxaRE
0
0
0
25
Together with @EverpureData, we built a two-node + witness setup for edge K8s HA. Two nodes run your containers and VMs, Portworx keeps their data in sync, and a lightweight witness casts the third quorum vote without hosting app data.
More details 👉 okt.to/qcsjOC
More details 👉 okt.to/qcsjOC
0
0
2
54
Choosing an endpoint is only the starting point. (Pun intended 😉)
A router can pick the endpoint, but someone still has to operate what sits behind it. That’s why you need a control plane for hybrid AI infrastructure.
Our CEO Tenry Fu explains why 👉 okt.to/6QTI5N
A router can pick the endpoint, but someone still has to operate what sits behind it. That’s why you need a control plane for hybrid AI infrastructure.
Our CEO Tenry Fu explains why 👉 okt.to/6QTI5N
1
0
2
60
Did you know we're an AWS Manufacturing and Industrial Competency Partner? ⚙️
Well, you do now. And that makes five AWS competencies for us, alongside Retail, Consumer Goods, Containers and Government.
See everything we do with AWS 👉 okt.to/ERv8Zi
Well, you do now. And that makes five AWS competencies for us, alongside Retail, Consumer Goods, Containers and Government.
See everything we do with AWS 👉 okt.to/ERv8Zi
0
0
1
30
Local inference is taking over Gibson Bar for #NVIDIAGTC Berlin.
If your token bill is getting out of hand, come talk to our team at our hospitality suite, just 4 mins from STATION-Berlin. There'll be live demos and canapés!
See our plans & book a 1:1 👉 okt.to/hKu81B
If your token bill is getting out of hand, come talk to our team at our hospitality suite, just 4 mins from STATION-Berlin. There'll be live demos and canapés!
See our plans & book a 1:1 👉 okt.to/hKu81B
0
0
0
46
We are happy to join @nvidia in today's launch of NVIDIA Run:ai as an open-source platform!
Together, we help customers put more of their GPU capacity to work without taking on greater operational complexity.
How it works and how to try it 👉 okt.to/PFbgOk
Together, we help customers put more of their GPU capacity to work without taking on greater operational complexity.
How it works and how to try it 👉 okt.to/PFbgOk
0
0
4
118
When inference runs on your own AMD Instinct GPUs, you can set every request to follow your routing policy, meter every token by team and use case, and use quotas to keep adoption from outrunning your controls.
Learn more about AMD Instinct™ Coder at okt.to/uzhdXZ
Learn more about AMD Instinct™ Coder at okt.to/uzhdXZ
0
0
0
55
We’re looking for a Senior Field Marketing Manager! 📣
This one’s for the event marketer who stays calm when plans change onsite and still has the ROI report ready once the show wraps.
If that sounds like you, check out 👉 okt.to/mYhlSW
This one’s for the event marketer who stays calm when plans change onsite and still has the ROI report ready once the show wraps.
If that sounds like you, check out 👉 okt.to/mYhlSW
0
0
0
47
PSA: we’re hiring! 📣
Open roles include Sr. Field Marketing Manager, Technology Alliances Manager, Accounts Payable Manager, Sr. Forward Deployed SE and Automation Engineer.
If one sounds like you (or maybe a friend?), check out our careers page 👉 okt.to/pumR9l
Open roles include Sr. Field Marketing Manager, Technology Alliances Manager, Accounts Payable Manager, Sr. Forward Deployed SE and Automation Engineer.
If one sounds like you (or maybe a friend?), check out our careers page 👉 okt.to/pumR9l
1
0
0
73
A WAN outage shouldn't close the checkout.
Our CTO Saad Malik joined Portworx by Everpure and IDC's Dave McCarthy on building a resilient retail edge, from VMs alongside containers to edge AI.
The recording's ready when you are 👉 okt.to/VLo64q
Our CTO Saad Malik joined Portworx by Everpure and IDC's Dave McCarthy on building a resilient retail edge, from VMs alongside containers to edge AI.
The recording's ready when you are 👉 okt.to/VLo64q
0
0
0
54
A single-destination AI strategy is getting harder to defend.
Agents turn one task into many model calls, and our own monthly token spend rose roughly 10x in six months.
And that's just 1 of 5 forces our CEO Tenry Fu lays out for hybrid AI in his blog 👉 okt.to/uQseSq
Agents turn one task into many model calls, and our own monthly token spend rose roughly 10x in six months.
And that's just 1 of 5 forces our CEO Tenry Fu lays out for hybrid AI in his blog 👉 okt.to/uQseSq
0
0
0
29
We're an NVIDIA AI Cloud Ready validated partner!
Getting AI factories into production fast is only the beginning, and PaletteAI provides the software foundation for running them in the years that follow.
More about the validation and why it matters 👉 okt.to/kJzpN7
Getting AI factories into production fast is only the beginning, and PaletteAI provides the software foundation for running them in the years that follow.
More about the validation and why it matters 👉 okt.to/kJzpN7
0
0
2
48