Andrey Cheptsov
@andrey_cheptsov
Followers
2K
Following
6K
Media
511
Statuses
9K
@dstackai. AI infra. Previously @JetBrains
Munich
Joined February 2008
This is our most impactful release so far. It brings more declarative GPU provisioning, much better observability and makes troubleshooting easier.
dstack 0.20 is out 🎉 A major end-of-year release towards GPU-native orchestration. Explicit provisioning control, better observability across system events, and more. Full breakdown:
0
2
5
Humbled to be featured by AMD ❤️ @AMD’s commitment to open-source and the developer ecosystem is exactly what this moment calls for. AI must be vendor-agnostic.
“You just use Docker, you just use vLLM, and it just works.” @andrey_cheptsov, Founder and CEO of @dstackai, explains why open standards and day-zero support matter and how AMD makes it easier for developers to build without vendor lock-in.
0
0
10
The most underrated impact of AI nobody talks about: Markdown going mainstream.
0
0
6
“You just use Docker, you just use vLLM, and it just works.” @andrey_cheptsov, Founder and CEO of @dstackai, explains why open standards and day-zero support matter and how AMD makes it easier for developers to build without vendor lock-in.
0
5
7
Calling Slurm users đź‘‹ If you work (or have worked) with Slurm, I have some questions about scheduling and training setups. Please DM or comment!
4
4
12
This is our most impactful release so far. It brings more declarative GPU provisioning, much better observability and makes troubleshooting easier.
dstack 0.20 is out 🎉 A major end-of-year release towards GPU-native orchestration. Explicit provisioning control, better observability across system events, and more. Full breakdown:
0
2
5
Calling Slurm users đź‘‹ If you work (or have worked) with Slurm, I have some questions about scheduling and training setups. Please DM or comment!
4
4
12
dstack 0.20 is out 🎉 A major end-of-year release towards GPU-native orchestration. Explicit provisioning control, better observability across system events, and more. Full breakdown:
dstack.ai
TBA
2
5
10
* @verdacloud (formerly DataCrunch) is partnering with dstack on GPU orchestration!🚀 The docs include a guide for using dstack to orchestrate GPU workloads: * Unified GPU provisioning and job scheduling * Support for dev, training, and inference workflows Docs:
1
2
7
@dstackai Psst 🤫 The Instant Clusters support is coming soon! Instant Clusters give you self-service and pay-as-you-go access to 16x-128x B200 GPUs with InfiniBand interconnect.
0
1
4
Introducing: a native integration for Verda x @dstackai, an open-source orchestrator built for AI 🧬 You can now manage Verda’s GPUs through dstack’s unified control plane - from dev to training to inference. Get started with our docs ➡️ https://t.co/XEQOQVUlLd
2
4
6
@HotAisle If you wait long enough, your competitor’s corpse eventually floats by. (After NVIDIA acquires it.)
0
1
4
Introducing Chatterbox Turbo, the fastest open source Voice AI model with emotions. Our gift to the dev community this holiday season! • ~6x faster RFT • Expressive sound tags: sighs, laughs, coughs • PerTh watermarking on every output Here's everything you need to know 👇
17
41
173
Do people actually use MCP servers in development? What real problems do they solve?
2
0
3
LMAO acquiring the companies that do open source isn’t “supporting” it
0
0
4
@andrey_cheptsov @dstackai How do I migrate to @dstackai and donate to keep you safe? I cannot see the doc so far :cry:
0
2
5