Evan Womack

In this conversation, @WEKA's Val Bercovici talks with @sdxcentral's Kat Sullivan about what’s changing and what teams should be paying attention to as they build for scale. Worth a watch if you’re thinking about where the bottlenecks are coming from 👇 spr.ly/6014BBky7E

Feels like AI M&A is shifting. Not just SaaS anymore—you’re seeing software + infrastructure come together. If tokens are the game, efficiency wins. That’s hard to do with one layer. This is exactly where @WEKA fits: spr.ly/6019BBTBFJ

As inference scales, more GPUs aren’t fixing things, they’re exposing the real issue. @WEKA is focused on what comes next: rethinking memory. Take a look 👇 spr.ly/6011BBua1D

The scale of AI infra investment right now is wild. $25B into @AnthropicAI $100B+ with @awscloud ~$200B total spend It’s not just about more GPUs anymore. It’s about efficiency. That shift is exactly where @WEKA plays. spr.ly/6017BBOYWc

Looking forward to this session with @nvidia's John Fragalla and @WEKA's Shimon Ben-David on what it actually takes to design an AI Factory for modern inference. Will be a really solid discussion if you’re working in this space 👇 spr.ly/6016BBOZJa

One thing that keeps coming up in AI infra conversations: it’s not just about performance, it’s about visibility. Hard to optimize what you can’t see. This is a solid breakdown of how teams are getting better insight into their environments ⬇️ spr.ly/6010BBMxgQ

Feels like everyone’s standardized on Kubernetes for AI, but storage is still where things get tricky. @WEKA is digging into this at #MilipolTechX 2026—should be a great session on what’s actually happening under the hood. 🗓️ spr.ly/6013BBMWZD

This session with @WEKA and RE:AI (Singtel) is a fantastic addition to the #MilipolTechX 2026 lineup. Should be a great conversation on what it really takes to move from infrastructure to intelligence. 🗓️ spr.ly/6018BBzlgK

🙌 Looking forward to this session at #MilipolTechX @WEKA’s Callan Fox will break down what’s actually slowing AI inference down, from time-to-first-token delays to KV cache I/O. 🗓️: spr.ly/6012BBz3MK

Do you actually know what AI infrastructure really costs? It’s easy to focus on GPUs and tokens, but a lot of the real cost sits elsewhere and there’s no single metric that tells the full story. Thought this was helpful insight from the team at WEKA ⬇️ spr.ly/6012BBz8EK

A lot of teams are hitting the same challenge. How do you take AI from pilots to real production without costs and performance getting out of control? Worth checking out if you are building in this space: spr.ly/6015BBJZ7H

At #NVIDIAGTC @WEKA’s Betsy Chernoff joined @Solidigm's Ace Stryker to break down how AI inference is shifting and why context memory and KV cache are becoming the real bottlenecks. Worth a read ⬇️ spr.ly/6015BBH7bV

Love seeing @WEKA recognized in the 2026 Storage 100 by @CRN 👏 One of the 50 coolest software-defined storage vendors! Check out the list ⬇️ spr.ly/6013BBGTbH

The infra that got AI through training doesn’t hold up for inference. At @WEKA, we’re focused on helping teams adapt where workloads are more memory-bound, KV cache drives cost, and cloud strategy is shifting. This breaks it down 👇 spr.ly/6014BBE00g

If Kubernetes workloads feel “stuck,” it’s often not CPU—it’s storage. @WEKA breaks it down in the blog post below. spr.ly/6012BBDbfy

🎉 Great week for AI and @weka at #HumanX in San Francisco!

Some great takeaways from @weka's Val Bercovici at #HumanX 👇 AI procurement is breaking, agents are becoming “employees,” and token efficiency is now critical.

Great insights from @WEKA's Shimon Ben-David during @VentureBeat’s AI Impact Tour on the “memory wall” and how to solve it. Take a look ⬇️ spr.ly/6012B6hqfC

Great things from Day 2 at #HumanX! Stop by @weka’s booth tomorrow and say hi if you’re at the show.

🎉 Congrats to our partner Firmus Technologies on $505M raised! WEKA is proud to support their energy-efficient AI factories across APAC. Get the full story: spr.ly/6014B67DkK