Sabitlenmiş Tweet
dalin stone
1.2K posts
dalin stone
@xdalinstone
jesus is king. i push buttons, things unbreak. fde @daytonaio
Idaho Katılım Nisan 2025
525 Takip Edilen128 Takipçiler
@yacineMTB one of the core maintainers of linux (or old maintainer) was an aenesthetist (?) I believe
English
@marccampbell @daytonaio should be fixed after one more PR approval 🙏🏼
English
@daytonaio y'all seem to have a really old version of openclaw (2026.2.1) preinstalled in your default images. thoughts on either upgrading it or just removing it?
my bootstraps have to work around this by uninstalling each time since i can't reliably figure out how to upgrade that install
English
English
@xdalinstone @NeuralAA @tan_stack @daytonaio @tannerlinsley Here are teh docs on code mode, i think it'll be clear after reading this:
tanstack.com/ai/latest/docs…
English
English
@NeuralAA @AlemTuzlak @tan_stack @daytonaio @tannerlinsley @AlemTuzlak I think you may be right. Let me see if there's another way. I guess there isn't some endpoint like '*.tanstack.com or tanstack.com*' that needs to be accessed right?
English
@AlemTuzlak @xdalinstone @tan_stack @daytonaio @tannerlinsley I may have done something without doing anything
GIF
English
@AlemTuzlak @NeuralAA @tan_stack @daytonaio @tannerlinsley Does adding it to our whitelist solve this?
English
English
infra is the WoW auction house rn btw. we live in amazing times.
English
dalin stone retweetledi
dalin stone retweetledi
.@mattturck and our CEO @ivanburazin chat on the MAD podcast
Matt Turck@mattturck
Every agent needs its own computer: my conversation with @ivanburazin, CEO of @daytonaio, about sandboxes and the emerging agent stack. 00:00 Intro 02:13 What is an AI agent sandbox? 03:17 Security risks of running agents locally 05:17 Stateful vs. stateless hyperscalers 07:04 The history of cloud IDEs and the end of localhost 09:45 Do all AI agents need a sandbox? 12:26 Sandbox use cases: RL evals & background agents 14:10 Unpacking the emerging AI Agent Stack 16:20 The unsolved problem of agent memory and learning 19:37 Where sandboxes fit in the agent harness 21:35 OpenAI, Anthropic, and agent SDKs 23:06 Ivan's founder journey: From CodeAnywhere to Daytona 26:59 GTM strategies and building developer communities 33:48 Why customer support is your best GTM strategy 35:34 Leveraging Twitter during the AI super cycle 40:50 The technical anatomy of a sandbox 41:53 Why fast spin-up speeds maximize GPU efficiency 46:09 Firecracker, QEMU, and isolation primitives 49:58 Why sandbox snapshots and state forking matter 51:40 Why Daytona built a custom scheduler from scratch 55:24 The challenge of long-running stateful sandboxes 58:10 The build your own sandbox trap 1:01:03 Why AI agents might trigger a global CPU shortage 1:02:46 The future of the AI Agent Stack
English
dalin stone retweetledi
Every agent needs its own computer: my conversation with @ivanburazin, CEO of @daytonaio, about sandboxes and the emerging agent stack.
00:00 Intro
02:13 What is an AI agent sandbox?
03:17 Security risks of running agents locally
05:17 Stateful vs. stateless hyperscalers
07:04 The history of cloud IDEs and the end of localhost
09:45 Do all AI agents need a sandbox?
12:26 Sandbox use cases: RL evals & background agents
14:10 Unpacking the emerging AI Agent Stack
16:20 The unsolved problem of agent memory and learning
19:37 Where sandboxes fit in the agent harness
21:35 OpenAI, Anthropic, and agent SDKs
23:06 Ivan's founder journey: From CodeAnywhere to Daytona
26:59 GTM strategies and building developer communities
33:48 Why customer support is your best GTM strategy
35:34 Leveraging Twitter during the AI super cycle
40:50 The technical anatomy of a sandbox
41:53 Why fast spin-up speeds maximize GPU efficiency
46:09 Firecracker, QEMU, and isolation primitives
49:58 Why sandbox snapshots and state forking matter
51:40 Why Daytona built a custom scheduler from scratch
55:24 The challenge of long-running stateful sandboxes
58:10 The build your own sandbox trap
1:01:03 Why AI agents might trigger a global CPU shortage
1:02:46 The future of the AI Agent Stack
English
dalin stone retweetledi
Loved to finally do a podcast with @mattturck
Matt Turck@mattturck
Every agent needs its own computer: my conversation with @ivanburazin, CEO of @daytonaio, about sandboxes and the emerging agent stack. 00:00 Intro 02:13 What is an AI agent sandbox? 03:17 Security risks of running agents locally 05:17 Stateful vs. stateless hyperscalers 07:04 The history of cloud IDEs and the end of localhost 09:45 Do all AI agents need a sandbox? 12:26 Sandbox use cases: RL evals & background agents 14:10 Unpacking the emerging AI Agent Stack 16:20 The unsolved problem of agent memory and learning 19:37 Where sandboxes fit in the agent harness 21:35 OpenAI, Anthropic, and agent SDKs 23:06 Ivan's founder journey: From CodeAnywhere to Daytona 26:59 GTM strategies and building developer communities 33:48 Why customer support is your best GTM strategy 35:34 Leveraging Twitter during the AI super cycle 40:50 The technical anatomy of a sandbox 41:53 Why fast spin-up speeds maximize GPU efficiency 46:09 Firecracker, QEMU, and isolation primitives 49:58 Why sandbox snapshots and state forking matter 51:40 Why Daytona built a custom scheduler from scratch 55:24 The challenge of long-running stateful sandboxes 58:10 The build your own sandbox trap 1:01:03 Why AI agents might trigger a global CPU shortage 1:02:46 The future of the AI Agent Stack
English
dalin stone retweetledi
dalin stone retweetledi
dalin stone retweetledi
dalin stone retweetledi
Announcing Cofounder 2: Run an entire company with agents.
It's the infrastructure for the one person billion dollar company - orchestrating agents across engineering, sales, marketing, ops, and design.
(and yes that's my real grandma in the video)
English
software engineers (who know what they're doing) running agents
Cormac@cormachayden_
software engineers before vs after agents
English
@ibuildthecloud @ivanburazin Are you asking if it possible to allow outbound traffic from a sandbox only to another sandbox? (I want no outbound traffic from A, EXCEPT FOR if outbound traffic is going directly to B and B is another sandbox)
English
@ivanburazin Would it be possible to only allow outbound traffic to another sandbox? The only ingress I see is preview URLs so I'm guessing sandbox to sandbox traffic is not a real use case because then you'd be routing through that proxy layer. Which is fine, just curious what is available.
English
Sandboxes need output network firewalling. Typically we thing mostly about ingress for firewall. We care about egress now.
English
dalin stone retweetledi
We spin up sandboxes in half the time it takes you to blink. 60ms is achieved by keeping it local.
We use the CPU, RAM, and hard disk of the underlying node. No external storage attachment (that most sandbox providers do with their VMs)
Snapshots are preloaded on the NVMes of the node. When you request a sandbox, it goes to that node and spins up from the same NVMe.
Can't fit every snapshot on every node, so we have an algorithm that sorts them across nodes. Plus snapshot layering and compression.
Zero network latency since everything's on the same node.
English
dalin stone retweetledi
I'm very not pro-hiring because I don't want to have 100 people and show that as a vanity metric.
We've closed Fortune 500s and enormous enterprises without a dedicated sales team.
Fewer people almost automatically mean less complexity within the team (we all still like each other very much)
But the amount of inbound we get now can't be solved without adding headcount.
I finally feel the need to add more AEs, growth people to own the entire channel from awareness to activation, and so on.
English