Skip to content

Quick Start

Go to chinilla.com/workspace in your browser.

The launcher gives you three options:

  1. Guided Wizard - Answer a few questions (domain, workflow, scale, concerns) and review an AI-generated first draft
  2. Template - Start from a pre-built system. Three categories:
    • Lessons (free) - Small, focused templates that teach specific behaviors (Pipeline Basics, Branching & Merging, Resilience Patterns, Batch & Scale, Full Loop, Circuit Breaker Test)
    • Examples (free) - Real-world workflows (AI Evaluation Pipeline, Pottery Business)
    • Interview (Pro) - Classic system design problems (URL Shortener, Chat Application, Notification System, Rate Limiter, File Upload Service, Pub/Sub Messaging, Content Delivery Network, News Feed / Timeline)
  3. Blank Canvas - Drop components manually

First time? Start with a Lesson template. Each component has a detailed description explaining what its behavior does and what to watch for when you run the simulation.

  1. Press A or click the Add button to open the component palette
  2. Drag a Client and an API onto the canvas
  3. Press C or select the Connect tool, then click the Client followed by the API
  4. Double-click a component to edit its name, description, and metrics

Click Play in the timeline bar to run the deterministic runtime. Watch packets flow through your connections, see where queues build up, and spot the first bottleneck. Live stats (health, delivery rate, packet counts) appear in the top bar during playback. Open the Overview panel from the left toolbar for deeper analysis, stability (Pro), and Monte Carlo (Pro).

Open the Stress Test panel from the Overview panel to push your design harder:

  • Quick Scenarios - One-click presets: Baseline, Peak Traffic, Outage, Slowdown, Flash Spike, Cascade
  • Traffic controls - Adjust seed packets (1-1000) in the timeline bar and traffic multiplier (1-10x) in the stress panel
  • Fault injection - Kill a component or inject latency to simulate degraded performance

Use the AI chat (H key) to ask for improvements, or manually add components to address bottlenecks. Re-run the simulation to confirm progress.