Giving Agents Computers — Ivan Burazin, Daytona
Get the full intelligence
Search transcripts, export clips, track mentions, and explore all topics from “Giving Agents Computers — Ivan Burazin, Daytona” inside PodZeus.
Ivan Burazin, CEO of Daytona, reveals how his company evolved from a browser-based IDE pioneer into the leading provider of composable, stateful sandboxes for AI agents—essentially the 'laptop for agents.' The pivot, driven by a pivotal demo of OpenDevin (now OpenHands), exposed a critical gap: human-focused infrastructure doesn't work for agents. What agents need is a persistent, fast, and dynamically resizable computer—like a Linux, Windows, or macOS machine that can be paused, resumed, and scaled instantly. Daytona’s bare-metal architecture, custom scheduler, and preloaded snapshots deliver sub-60ms spin-up times and support 500,000+ concurrent CPUs daily. The real breakthrough? Agents aren’t just running code—they’re automating legacy enterprise workflows in locked-down Windows apps, a market worth $10 trillion. Burazin argues that the future of AI infrastructure isn’t AWS or Heroku, but a Stripe-like API-first platform built specifically for agents, where compute is as composable and programmable as a database or payment processor. And the most powerful differentiator? A 25-person team that responds to Slack messages in minutes, not days.
Agents need composable computers, not just VMs—Daytona provides stateful, resizable, bare-metal sandboxes with sub-60ms spin-up times.
The market for AI agent infrastructure is exploding, growing 74% month-over-month, with demand driven by automation of legacy enterprise systems.
The biggest opportunity lies in automating Windows-based legacy apps—Burazin estimates a $10 trillion market for agent-driven RPA.
Daytona’s open-core model with AGPL 3.0 licensing protects against competitors while enabling enterprise adoption through transparency.
Responsiveness is the #1 sales differentiator: customers cite 5-minute Slack response times as the reason they choose Daytona over competitors.
…and 3 more takeaways available in PodZeus
From Code Anywhere to Daytona: A Founding Story
Ivan Burazin recounts his journey from co-founding Code Anywhere, the first browser-based IDE, to launching Daytona. He reflects on the early days of cloud coding, the challenges of building infrastructure from scratch before Docker and Kubernetes, and how the experience shaped Daytona’s architecture.
The Pivot That Changed Everything
“I've never experienced, I've done multiple companies in my life, I've never experienced this, that people literally call you if you do not give them access. Like, they want access right now.”
Why Bare Metal and a Custom Scheduler?
Daytona runs on bare metal with its own scheduler, enabling sub-60ms spin-up times and stateful resumption. This architecture allows for instant snapshot loading and eliminates network latency, making it ideal for agents that need to pause and resume like humans.
The $10 Trillion Windows Agent Opportunity
“If even a startup like ours and using all the hottest tools still needs a computer agent, what hope does, you know, Goldman have to have a headless, right?”
The Spiky Workload Problem
Unlike human workloads, agent workloads are highly spiky—sudden surges of 100,000 CPUs in seconds. This creates a new infrastructure challenge: how to handle bursts without over-provisioning or sacrificing speed.
“I've never experienced, I've done multiple companies in my life, I've never experienced this, that people literally call you if you do not give them access. Like, they want access right now.”
“If even a startup like ours and using all the hottest tools still needs a computer agent, what hope does, you know, Goldman have to have a headless, right?”
“There will be an equivalent, everyone says like an AWS for AI agents, but your answer, like it might look more like Stripe than AWS in a sense.”
Host
Guest
Daytona
organization
Ivan Burazin
person
Code Anywhere
organization
OpenDevin
product
GitHub
organization
OpenHands
product
Anthropic
organization
Vercel
organization
OpenAI
organization
Microsoft Power Automate
product
Moonlake: Causal World Models should be Multimodal, Interactive, and Efficient — with Chris Manning and Fan-yun Sun
Latent Space: The AI Engineer Podcast • 1h 6m • 4/2/2026
Marc Andreessen introspects on The Death of the Browser, Pi + OpenClaw, and Why "This Time Is Different"
Latent Space: The AI Engineer Podcast • 1h 16m • 4/3/2026
Extreme Harness Engineering for Token Billionaires: 1M LOC, 1B toks/day, 0% human code, 0% human review — Ryan Lopopolo, OpenAI Frontier & Symphony
Latent Space: The AI Engineer Podcast • 1h 12m • 4/7/2026
Notion’s Token Town: 5 Rebuilds, 100+ Tools, MCP vs CLIs and the Software Factory Future — Simon Last & Sarah Sachs of Notion
Latent Space: The AI Engineer Podcast • 1h 17m • 4/15/2026
🔬 Training Transformers to solve 95% failure rate of Cancer Trials — Ron Alfa & Daniel Bear, Noetik
Latent Space: The AI Engineer Podcast • 1h 25m • 4/20/2026
Get the full intelligence
Search transcripts, export clips, track mentions, and explore all topics from “Giving Agents Computers — Ivan Burazin, Daytona” inside PodZeus.
Start discovering podcast insights today
Start with a 7-day trial and explore a growing catalog of popular podcasts. No credit card required.
No credit card required • 7-day trial • Cancel anytime