JoanMedia

Hello masters, I am currently having an amazing experience talking at the AutomationStar2025. Thanks always for your support and I hope to see you soon in a new video 🎥

Skip the Carousel

1 2 3 4

5 months ago | [YT] | 7

JoanMedia

🎯 Playwright + TypeScript + AI: now live on Udemy

I wanted to create something concise and practical, not another course filled with commands and theory.

This one is all about building your own Playwright framework just like we do in real QA projects:
- Page Objects, fixtures, secrets, and test data
- Parallel runs and CI/CD setup
- Integrations with AI tools (MCP servers, Cursor, Copilot) to speed up test creation and maintenance

You’ll end up with a ready-to-use, real-world framework you can adapt to any project.

Plus, I added a few extra lectures on API testing, mocking, and accessibility 💡
If you're looking for a quick, useful, and hands-on course, this is for you.

www.udemy.com/course/playwright-typescript-framewo…

Coupon code $9.99: 1124FE43797001D75328

6 months ago | [YT] | 4

JoanMedia

🎁 I’m building a new Playwright course on Udemy, focused on creating a real testing framework from the ground up using Page Objects, fixtures, data management, and CI integration.

If you’ve been meaning to level up your Playwright skills or understand how to structure a framework that scales, this one’s for you.

I’ll gift free access to the first 15 people who register here 👇

🎁 lnkd.in/eU7JCcB3

Leave your email there and I’ll reach out when the course drops.

#Playwright #Testing #QA #Udemy

6 months ago | [YT] | 5

JoanMedia

🧭 Testing what your users actually see, not just what your APIs respond.



That’s what the k6 Browser module is all about.

It goes beyond protocol-level load testing (requests per second, latency, etc.) and opens a real browser, letting you measure frontend performance the same way a user experiences it.



So now, instead of only asking “How fast does my API respond?”, you can ask:

- How long until the page becomes interactive?

- Are spinners taking too long to disappear?

- What happens to the frontend when I push heavy backend load?



It’s still k6, so you get all the same scripting power and metrics, but now with browser-level visibility:



✅ Web Vitals (LCP, FCP, TTFB, CLS)

✅ Real page rendering times

✅ Element interaction checks



Think of it as combining your load test and user experience test in one run — backend stress meets frontend reality.

6 months ago | [YT] | 3

JoanMedia

🧠 SharedArray in ‪@Grafana‬ K6

When you run a load test, each virtual user (VU) works independently, loading the same data repeatedly. That’s a lot of wasted memory.

SharedArray fixes that. It loads your data once, stores it in memory, and allows all VUs to use it. Same data, no repetition.

Why it matters:

- Faster test start

- Less memory use

- Cleaner setup


It’s like everyone in a kitchen using the same fridge instead of bringing their own groceries 🍕

6 months ago | [YT] | 4

JoanMedia

Hey everyone 👋
No new video this week — I’ve been feeling a bit burnt out and need to take a short break.
Hopefully next week I’ll be back with a new one.
Love you all ❤️

6 months ago | [YT] | 3

JoanMedia

🔄 The Test Lifecycle in k6


Every k6 test follows the same sequence of stages:

1️⃣ Init: prepare the script (imports, options, globals). Required.

2️⃣ Setup: create data or prepare the test environment. Optional.

3️⃣ VU code: the main test logic, executed by Virtual Users. Required.

4️⃣ Teardown: clean up and process results at the end. Optional.


💡 Think of it like a journey:

Init is packing your bag 🎒

Setup is checking into the hotel 🏨

VU code is living the actual trip ✈️

Teardown is closing the trip and heading home 🏠



With this structure, your tests stay organized and predictable, and you know exactly where to add preparation, execution, or cleanup code.

6 months ago | [YT] | 2

JoanMedia

✅ Checks in k6

Checks are validations inside your test. They answer questions like:

- Did the response come back with status 200?

- Does the body contain the expected text?

- Is the size of the response correct?



Unlike thresholds, a failed check does not stop the test.

The script keeps running, and k6 simply records how many checks passed or failed.



👉 Think of checks as quality stamps on each request. Some may pass ✅, others may fail ⚠️, but the assembly line (your test) keeps moving.



Note: If you want a failed check to actually break the test, combine it with a threshold. That’s how you turn validations into automated pass/fail gates for CI pipelines.

6 months ago | [YT] | 1

JoanMedia

✅❌ Thresholds in k6


A threshold is a rule that decides if your test passes or fails.

It’s like saying: “If my system is slower than this, it fails.”


Why they matter:

- They make your performance goals (SLOs) clear.

- They let you automate: if the system is too slow or errors too much, the test stops with a fail.

- They save you time: you only check results when something goes wrong.


Examples of thresholds:

- Less than 1% errors.

- 95% of requests respond under 200ms.

- A key endpoint always under 300ms.


💡 Think of it like a speed limit 🚦. If you stay under, all good. If you go over, the system is “caught speeding” and the test fails.

7 months ago | [YT] | 4

JoanMedia

What do you want to learn next on the channel? 🤔

7 months ago | [YT] | 0