Move Over Selenium: Using Playwright in Ruby

The Selenium Struggle

If you’ve done any browser automation in Ruby, you know the "Selenium Dance."

1. You try to click a button.
2. The test fails because the element wasn't "ready" yet.
3. You add sleep(2).
4. It works for a week, then fails again.
5. You add wait.until { ... } and your code becomes a mess of nested blocks.

Selenium was built for a different era of the web—an era before heavy React/Vue apps and complex asynchronous state. It requires a clunky "WebDriver" middleman to talk to the browser, which adds latency and instability.

It’s time to move over. Playwright has arrived in the Ruby ecosystem, and it changes everything.

---

Why Playwright Wins

Playwright (developed by Microsoft) was designed for the modern web. In the Ruby world, we use the excellent playwright-ruby-client gem. Here is why it is superior:

1. Auto-Waiting (The "Sleep" Killer)

Playwright performs "actionability checks" before every interaction. If you tell it to click a button, it automatically waits for that button to be visible, enabled, and stop moving before it tries to click. You can finally delete all your sleep calls.

2. Fast Network Interception

Playwright allows you to see, modify, or block network requests. As we discussed in previous articles about internal APIs, this is a superpower. You can tell Playwright to ignore all CSS and Image requests to make your scraper 5x faster.

3. Multiple Contexts

You can open 10 "Browser Contexts" within a single browser instance. These are like completely isolated incognito windows. This is perfect for testing different user roles or scraping sites that require multiple separate sessions without the overhead of booting 10 browsers.

---

Getting Started

First, add the gem to your Gemfile:

gem 'playwright-ruby-client'

Then, install the browsers (this is a one-time setup):

bundle install
bundle exec playwright install

---

The Basic Script

Here is how clean a Playwright script looks compared to the verbose Selenium setup:

require 'playwright'

Playwright.create(playwright_cli_executable_path: './bin/playwright') do |playwright|
  # Launch Chromium (or firefox / webkit)
  playwright.chromium.launch(headless: true) do |browser|
    page = browser.new_page
    page.goto('https://github.com/login')

    # No 'sleep' needed! Playwright waits for the input to appear.
    page.fill('input[name="login"]', 'your_username')
    page.fill('input[name="password"]', 'your_password')
    page.click('input[type="submit"]')

    puts "Logged in as: #{page.title}"
  end
end

---

Killer Feature: Codegen

One of the best things about Playwright is Codegen. You can run a command that opens a browser, records your clicks, and generates the Ruby code for you.

bundle exec playwright codegen https://example.com

As you click around, the terminal will live-stream the Ruby code required to replicate those actions. It’s the fastest way to build a scraper.

---

Advanced: Intercepting JSON

Remember how we talked about scraping internal APIs? Playwright makes this trivial. You can listen for the exact moment the website fetches its data:

page.on('response') do |response|
  if response.url.include?('/api/v1/products')
    puts "Caught the data: #{response.json}"
  end
end

page.goto('https://example.com/products')

---

Selenium vs. Playwright: The Comparison

Feature	Selenium	Playwright
Speed	Slow (HTTP overhead)	Fast (Direct Pipe)
Waiting	Manual / Brittle	Native Auto-waiting
Browsers	All (via WebDrivers)	Chromium, Firefox, WebKit
Mobile	Limited	Native Emulation
Network Control	Very Hard	First-class citizen

---

Summary

If you are starting a new project in 2026, there is almost no reason to choose Selenium. Playwright is faster, more reliable, and built for the way we build websites today.

It turns "browser automation" from a frustrating chore into a reliable part of your Ruby stack.

---

Have you made the switch to Playwright yet? What was the biggest "Aha!" moment for you? Let's discuss in the comments! 👇