Blog
Insights on browser automation, AI agents, and distributed computing
Frontend Dev Without a Browser
How we built a visual development workflow where an AI agent edits React code, builds to S3, and uses a remote browser API to see what it made — no Chrome installed.
Read more →E2E Testing Tips: Lessons from Testing with Riddle
Practical tips for browser automation and E2E testing learned from extensive use of Riddle's API. Sync vs async, handling large responses, debugging strategies, and cost optimization.
Read more →Teaching AI to Ski: A Browser Automation Adventure
How we used Riddle to train an AI agent to play a browser-based skiing game, combining vision models with real-time browser control.
Read more →Building Vision Agents with Browser Automation
A deep dive into creating AI agents that can see and interact with web pages using screenshots, combining LLMs with browser automation for intelligent web navigation.
Read more →The Economics of "Chatty" Agents (Part I)
Browser-based agents burn money and time when structured wrong. Learn how batching browser work into sessions saves both. Part I focuses on the economics and architecture.
Read more →Your Agent Doesn't Need to See Every Step (Part II)
When is it safe to batch browser actions? How do you handle failures mid-script? Part II covers the heuristics for smart batching without losing debuggability.
Read more →