WEKA

NeuralMesh™ by WEKA® - The world's only storage system purpose-built for AI. Accelerate performance, deploy anywhere, grow stronger with scale.

[WEKA @weka]

⏪ ICYMI: Val Bercovici shared his perspective with @VentureBeat on why the AI race is shifting from models to infrastructure efficiency.

A timely look at how DeepSeek is reshaping AI economics with lower inference and cache-read costs.

Take a look 👇 https://t.co/MNScxqna0M

[WEKA @weka]

AI has moved beyond experimentation. Now the challenge is inference economics: running AI at scale, supporting agentic workloads, and meeting sovereignty requirements without runaway costs.

See how WEKA is tackling it at @superai_conf next week.

🗓️ https://t.co/10JVKEU3Zc

[WEKA @weka]

💭 What if data sovereignty is an architectural decision, not a compliance checkbox?

At #ISC26, Chris Ott (WEKA) and Ali Syed (DCAI) will share lessons from building Gefion, one of Europe’s first sovereign AI supercomputers.

🔗 https://t.co/ebNsmTLFhh

[WEKA @weka]

Legacy vendors will confidently tell you they can handle your AI workload. Then they go check the back. You've seen the back.

Nobody wants to go there.

🚪 https://t.co/1Kd0PdiFRT

[WEKA @weka]

🤝 Through our partnership with @Yoodli, WEKA is enabling go-to-market teams to practice real-world sales scenarios, receive objective feedback, and demonstrate readiness through scalable certification.

Here's how 👇 https://t.co/D3e4GowB4O

[WEKA @weka]

⚡ Inference is becoming the economic center of AI.

At #SuperAI 2026, WEKA is spotlighting how organizations can break through the AI memory wall, optimize token economics, and build infrastructure for cost-efficient AI at scale.

🔗 https://t.co/7Ydn4N8Pvu

[WEKA @weka]

📰 WEKA’s Val Bercovici puts it plainly: the AI race is no longer just about models — it’s about infrastructure efficiency.

Learn more in this article 👇 https://t.co/99bUSCmWij

[WEKA @weka]

Seismic simulations, fusion research, and grid studies demand massive throughput and nonstop GPU utilization at scale.

NeuralMesh eliminates storage bottlenecks to accelerate energy AI innovation.

Here's how 👇 https://t.co/qg2Vlu2f7H

[WEKA @weka]

⚡ AI and HPC workloads are converging fast, and infrastructure has to keep up.

Join WEKA’s @shimonbd at #ISC2026 for a live look at building high-performance infrastructure for accelerated compute at scale.

🔗 https://t.co/HK2N1HqKpB

[WEKA @weka]

@superai_conf @RaiseSummit 🇸🇬 SuperAI: https://t.co/2v6SuAB6HS
🇩🇪 ISC: https://t.co/1VexEX0BLr
🇫🇷 RAISE Summit: https://t.co/EfFkkfN0Yy

[WEKA @weka]

🛫 Next stop: Singapore, Hamburg, and Paris.

We’re hitting the road to talk inference economics, memory efficiency, sovereign AI, and accelerated compute at scale.

🇸🇬 @superai_conf | June 10–11
🇩🇪 ISC | June 23–25
🇫🇷 @RAISESummit | July 7–9

👇 Full event details below

[WEKA @weka]

Today, we honor and remember the brave service members who made the ultimate sacrifice in service to our country.

As we spend time with family and friends this #MemorialDay, we pause to reflect on the courage, service, and sacrifice that made those moments possible.

We remember. 🇺🇸

[WEKA @weka]

🎤 Val Bercovici joined Hope King, Jonah Surkes, and Dan Lawrence at #HumanX to discuss the real economics behind AI at scale.

As token demand and energy constraints rise, infrastructure efficiency is becoming critical to AI performance and cost.

💬 https://t.co/Sg7jOJ6KWt

[WEKA @weka]

🤝 Excited to collaborate with @Nokia in its new AI Networking Innovation Lab alongside other leaders shaping the future of AI-native infrastructure.

As AI workloads scale, collaboration across networking, memory, storage, and compute is more critical than ever.

Full story ⬇️ https://t.co/waqqINbQB5

[WEKA @weka]

AI multitenancy shouldn’t force tradeoffs between isolation, performance, and efficiency.

NeuralMesh supports both physical and virtual isolation models from one platform and control plane — enabling true isolation with zero waste.

Here's how 👇 https://t.co/m537KauYPV

[WEKA @weka]

AI isn’t getting cheaper — token demand is exploding.

At #HumanX, WEKA's Val Bercovici joined industry leaders to discuss token ROI, the AI memory wall, and why infrastructure efficiency now defines AI economics.

Watch the full conversation 👇 https://t.co/RzLGvFuOUs

[WEKA @weka]

🏭 Manufacturing AI only works if infrastructure can keep up.

NeuralMesh helps accelerate predictive maintenance, robotics, and defect detection with ultra-low latency and real-time GPU performance.

Here's how 👇 https://t.co/6y7250aGSC

[WEKA @weka]

🌐 The global AI memory shortage is already reshaping how organizations scale #AIInfrastructure.

In @Forbes, WEKA CEO @liranzvibel explains why inference demand is colliding with limited memory supply and why optimization—not stockpiling hardware—is the path forward. https://t.co/tC1mc5zqw4

[WEKA @weka]

@superai_conf We can't wait!

[WEKA @weka]

🇩🇪 We’re heading to #ISC26 to showcase how NeuralMesh powers simulation, AI training, inference, and automation together on a single platform built for accelerated compute.

👋 If you’re attending the show, stop by and connect with the WEKA team. https://t.co/uBeCBCsjFY