cloudflareJan 29, 2026

Introducing Moltworker: a self-hosted personal AI agent, minus the minis (opens in new tab)

Moltworker adapts the self-hosted Moltbot personal AI assistant to run on Cloudflare without requiring users to buy a dedicated Mac mini. It combines a Cloudflare Worker, Sandbox SDK, Browser Rendering, R2, AI Gateway, and Cloudflare Access to provide a globally available, secured deployment. The result is a managed infrastructure layer around Moltbot’s standard Gateway runtime while preserving its integrations and persistent state. ## Running a Personal Agent on Cloudflare - Cloudflare Workers increasingly supports Node.js APIs natively, reducing the need for compatibility hacks and making it easier to run existing JavaScript and TypeScript packages. - An internal test of the 1,000 most popular NPM packages found that only 15 relevant packages failed to run in Workers. - Although much of Moltbot runs inside a container, improved Workers compatibility is useful for building agent logic closer to users. - Cloudflare’s Developer Platform provides the main infrastructure components: - **Sandboxes** for securely running untrusted code. - **Browser Rendering** for automated headless browser interactions. - **R2** for persistent object storage. - Cloudflare’s global network for scalability and security. ## Moltworker Architecture - Moltworker consists of: - An entrypoint Worker serving as an API router and proxy. - Cloudflare Access protecting the Worker and administration interface. - A Sandbox container running Moltbot’s standard Gateway and integrations. - R2 for persistent storage. - This structure separates the public API and administrative layer from the isolated environment where the agent executes. ## AI Gateway Integration - Cloudflare AI Gateway proxies requests between Moltbot and AI providers. - It provides: - Centralized request visibility. - Cost monitoring, logs, and analytics. - Provider and model switching without changing Moltbot code. - Fallback providers or models for improved reliability. - Secrets can be managed through: - **Bring Your Own Key (BYOK)**, where provider credentials are stored centrally. - **Unified Billing**, where users purchase credits and Cloudflare handles provider billing. - Integration requires creating an AI Gateway instance, enabling a provider such as Anthropic, and setting `ANTHROPIC_BASE_URL`; Moltbot itself does not need code changes. ## Sandbox-Based Execution - The Sandbox SDK runs agent code in isolated environments built on Cloudflare Containers. - It provides simplified APIs for: - Executing commands. - Managing files and directories. - Running background processes. - Exposing services. - Executing code in contexts such as Python. - The SDK abstracts container lifecycle, networking, filesystem, and process-management concerns behind TypeScript APIs. Moltworker offers a way to run a capable personal AI agent online with managed security, storage, browser automation, and model access—without maintaining dedicated hardware.

browser-renderingai-agentcloudflare-workersserverless+4

figmaMay 19, 2017

Reflecting on Figma’s First Year | Figma Blog (opens in new tab)

Figma’s first year focused on proving that a browser-based design tool could deliver both powerful graphics performance and trustworthy collaboration. The company moved from being dismissed as a “toy” to becoming a serious platform through rapid feature releases, performance improvements, and better support for existing Sketch workflows. Its central insight was that multiplayer editing creates not only collaboration, but also a shared source of truth for entire product teams. ## Building a Browser-Based Design Platform - Figma faced major technical challenges in rendering complex graphics smoothly in a browser. - It also had to persuade designers that a web-based tool could be reliable enough for professional work. - The team spent its first year speaking with designers, engineers, and product managers to understand the needs of growing organizations. - Figma invested heavily in: - Faster file loading and rendering - Workflow improvements and polish - Continued refinement of Sketch import - Because releases came so quickly, the company collected its major updates in one anniversary retrospective. ## Collaborative Design and Multiplayer Editing - Public multiplayer editing launched on September 28, 2016, making Figma the first design tool to offer simultaneous editing. - Designers initially reacted both enthusiastically and skeptically to the idea of “pair design.” - After a year, teams were using multiplayer editing to: - Ship products under tight deadlines - Mentor designers remotely - Conduct more efficient design critiques - The larger benefit was a single source of truth: - Files could be shared through a persistent URL. - Updates appeared at the same location without requiring teams to synchronize assets manually. - Teams avoided overwritten files and outdated versions. - Different roles could work directly with the design: - Copywriters could test messaging. - Developers could inspect design attributes for implementation. - Product managers could monitor progress without relying on Slack updates. ## Components and Design Systems - As design became more central to product development, companies needed systems for maintaining consistent UI elements across teams and files. - Figma introduced components on December 8, 2016, allowing designers to synchronize reusable elements such as navigation menus and status bars. - Component overrides let designers customize individual instances—for example, changing color or typography—without disconnecting them from the underlying component. - This balanced flexibility with consistency, allowing teams to benefit from a broader design system while still adapting individual designs. Figma’s first-year progress suggests that browser-based design succeeds when performance, compatibility, and collaboration are treated as equally important. Its persistent shared files and emerging design-system features aimed to make the design process more connected across the entire product team.

browser-renderingfigmadesign-systemscollaborative-design+2