Verda

Today we are announcing $189M in new funding and becoming Europe’s latest unicorn.

This funding accelerates what we’ve been doing from day 0:

Building the most reliable and performant AI infrastructure for our customers.

Read more in the thread 🧵

2,000 builders, 48 hours, and Verda credits to all competing teams. At @hackJunction 2026, we own the AI Infrastructure Track and we're the main partner alongside @arduino and @GoogleDeepMind. Here's what we're bringing:

1️⃣ Verda credits for all teams, solving any challenge
2️⃣ €10,000 credits for the winning team of our “Best use of Verda” bounty
3️⃣ Our engineers and talent team at the booth all weekend
4️⃣ The main-stage keynote and a hands-on workshop

Nov 13-15 | Hype Areena | Booth 6

Save the date, meet us at the booth, full plan here: verda.com/blog/junction-…

Verda Inference is built around @sgl_project and tuned by SGLang core contributors on our team.

Thanks to the SGLang community for making frontier-level inference open and helping us bring Kimi K3.1 Pro to Verda Inference on day 0.

Today we're introducing Verda Inference, in preview, with day-0 access to Kimi K3.1 Pro, in partnership with @Kimi_Moonshot 🧵

Frontier open-weight models through one API, engineered for every request, from a single prompt to long-horizon agents.

verda.com/blog/verda-inf…

Large training datasets can leave GPUs idle during distributed training.

Here’s how to prevent GPU starvation with data streaming, end-to-end on Verda.

See our example with .mds, a Slinky environment and our new S3-compatible Object Storage:

verda.com/blog/data-stre…