Members-Only
Recent Talks & Demos are for members only
You must be an AI Tinkerers active member to view these talks and demos.
Booting Any AI Coding Agent on a Disposable Cloud Machine in 2 Seconds
Launch AI coding agents like Claude Code and Codex on isolated cloud machines in seconds. Get a root shell, assign tasks, and reconnect later to see completed work, all without local installs.
I built Zipbox.ai, a platform that boots AI coding agents like Claude Code, Codex, Cursor, Grok, and others on disposable Firecracker microVMs in about two seconds.
No SDKs, API keys, or local installs. Pick an agent, get a root shell in an isolated cloud machine, enable YOLO mode if you want, give it a task, and walk away. The VM keeps running even if you close your browser.
In the demo, I’ll boot Claude Code, have it clone a repo, write tests, and run them. Then I’ll close the tab, come back a few minutes later, reconnect, and show the completed work. Every sandbox gets its own kernel, public HTTPS URL, and is billed by the second.
Zipbox boots AI coding agents inside disposable, hardware-isolated Firecracker microVMs.
- Firecracker microVMFirecracker is an open-source virtualization technology built by AWS that runs secure, multi-tenant workloads in lightweight virtual machines called microVMs.Developed by Amazon Web Services to power resource-intensive services like AWS Lambda and AWS Fargate, Firecracker is a minimalist virtual machine monitor (VMM) written in Rust (1.1.1, 1.1.6, 1.1.7). By utilizing the Linux Kernel-based Virtual Machine (KVM), it strips away legacy devices and non-essential guest functionality to establish a highly restricted attack surface (1.1.1). The result is a hypervisor capable of launching fully isolated microVMs in under 5 milliseconds while maintaining a tiny 5 MB memory footprint (1.1.1, 1.1.2). This design allows operators to achieve the strict security boundaries of traditional hardware virtualization alongside the rapid startup times and high density of standard container deployments (1.1.1, 1.1.2).
- BunBun is the fast, all-in-one JavaScript runtime and toolkit: it replaces Node.js, npm, and multiple development tools with a single, cohesive executable.Bun delivers a high-performance JavaScript runtime, built from scratch in Zig and powered by Apple's JavaScriptCore engine (not V8). It functions as a drop-in replacement for Node.js, but unifies the entire development workflow: the runtime, package manager (e.g., `bun install`), bundler (`bun build`), and test runner (`bun test`) are all integrated. This architecture dramatically reduces startup times and memory usage, offering significant speed gains, especially in tasks like package installation (up to 30x faster than npm) and server-side rendering.
- TypeScriptTypeScript is an open-source superset of JavaScript: it adds static typing and compiles to clean, standards-based JavaScript.TypeScript is a high-level, open-source language developed by Microsoft: it acts as a superset of JavaScript, adding a powerful static type system. This system enables compile-time type checking, catching errors before runtime (a critical benefit for large-scale applications). The TypeScript Compiler (TSC) reliably transpiles all code into clean, standards-based JavaScript (ES3 or newer), ensuring compatibility across any browser or host environment (Node.js, React.js, etc.).
- Claude CodeAnthropic's agentic coding tool: Unleash Claude's raw power directly in your terminal or IDE to turn complex, hours-long workflows into a single command.Claude Code is Anthropic’s powerful agentic coding assistant, designed for high-velocity development. It operates natively within your terminal, IDE (VS Code, JetBrains), or via a web interface, allowing you to delegate complex tasks like feature building, bug fixing, and codebase navigation. The agent plans, edits files, executes commands, and creates commits, maintaining awareness of your entire project structure. Internally, Anthropic engineers using Claude Code reported a 67% increase in productivity, demonstrating its capacity to deliver significant gains for Pro and Max plan users.
- CodexCodex is OpenAI's autonomous AI software engineering agent: it executes full development tasks in a sandboxed cloud environment.Codex is the advanced, cloud-based software engineering agent from OpenAI, built on a specialized model like `codex-1` (a fine-tuned version of `o3`). It operates on an asynchronous delegation model, allowing developers to assign complete tasks—not just receive suggestions—via the ChatGPT interface. The agent works independently in a secure, isolated cloud container provisioned with the user's GitHub repository and environment. It reads code, writes new features, fixes bugs, runs tests, and drafts Pull Requests (PRs) for review, significantly accelerating the development lifecycle. Access is provided through ChatGPT Plus, Pro, and Enterprise plans.
Related talks
More from the community
Cloud-Hosted, Multi-Agent Workspaces (or: Gas Town in the Sky)
Seattle
Soul, Memory, Skills, Brain: Building a Zero-PII Personal AI Agent in Docker
Austin
Jules: An AI Agent That Thinks, Argues, and Writes
Austin
Building a Digital Supermax: Architecting Air-Gapped Runtimes for Untrusted Agents
DC
Learn how to build secure, air-gapped runtimes for AI agents like OpenClaw. This talk explores network infrastructure and…
OpenClaw, Now With a Seatbelt: running agents in micro-VMs on your Mac
Toronto
Learn how OpenClaw agents run in isolated micro-VMs on Mac, with network and secret mediation, creating a robust…
Building Bob's Cabins Across Three Agent Tools Without Losing My Mind
Nashville
Learn how to build software across multiple AI tools without losing context. This talk demonstrates a local, shared…
Compose Email
Loading recent emails...