May at Verda: new silicon, new office, new research. We're deploying NVIDIA Rubin, new software updates and deepening key partnerships. Here's everything in one place.
Platform updates
Status page
Our status page at status.verda.com gives you real-time visibility into system health and upcoming maintenance windows. Subscribe to get automated notifications delivered straight to your inbox, so you'll always know what's happening on the platform before it affects your workloads.
Discontinuing HDD
We're discontinuing HDD-based services on Verda. UI controls have already been removed from the console, and we're in the process of removing HDD services from the API and SDK. Existing HDD data will be migrated to NVMe before HDD nodes are physically retired.
If you have data on HDD storage, our team will be in touch with migration details.
Confidential Computing
Confidential Computing is now available in self-service on Verda, running on RTX PRO 6000. Hardware-level isolation that protects your workloads even during processing, provisionable directly from the console. Get started: https://console.verda.com
Company News
NVIDIA Rubin
Verda will be among the first to deploy NVIDIA Vera Rubin NVL72 and HGX Rubin NVL8 at scale, with the earliest deployments starting in H2 2026 and expected capacity exceeding 100,000 GPUs throughout 2027. As an NVIDIA Preferred Partner, we're bringing first-availability across three regions in parallel: Europe, the US, and APAC.
Supermicro
Supermicro is one of the partners powering Verda's full-stack AI cloud, supplying the systems behind our latest NVIDIA Blackwell deployments, including GB300 NVL72, HGX B300, HGX B200, and RTX PRO 6000 racks. Their deep expertise in system design and liquid cooling helps us bring the latest hardware online quickly, sustainably, and at the density our customers need.
Research
Guide: NVFP4 GEMM Kernel for B200
Part 2 of our NVFP4 mini-series is live. We walk through building an NVFP4 GEMM kernel for B200 in CuTeDSL, covering tiling, swizzling, block-scaled FP4 quantization, and the persistent 2-CTA pipeline that ties it all together.
Events
MLSys 2026
We brought a booth to MLSys 2026 and had a blast. A few days of talking with researchers and engineers about training efficiency, inference, and where AI infrastructure is heading next. Huge thanks to everyone who swung by to chat. Until next year.
Upcoming Events
We're hosting a day-long GPU programming workshop in Helsinki with Erik Schultheis (GPU MODE), focused on the often-overlooked side of CUDA: memory-bound kernels. Participants will run code live on Verda's infrastructure, including RTX 6000, B200, and H200, and work through everything from memory hierarchy basics to warp-level programming and asynchronous data movement.
Job Openings
We’re growing fast and expanding our team. If you’re passionate about AI infrastructure and want to help build Europe’s next hyperscaler, we’d love to hear from you.
Explore our open roles and join us on the journey to shape the future of cloud computing:
- Product Manager - Serverless & Containers
- Design Lead Data Center
- Senior Data Center Operations Engineer
- Member of Technical Staff, AI Infrastructure
- Head of Data Center Operations
- ML Infrastructure Engineer, Forward-Deployed
- Senior ABM/ABX Campaign Manager