profile

Filip Hric

LLMs are not magic

Published about 1 year ago • 1 min read

Filip Hric

12th February 2025

LLMs ARE NOT MAGIC

Hello Reader,
Ever wondered why ChatGPT sometimes gives odd or completely wrong answers? My personal guess is that we tend to treat them like magic. Or treat them like a person. But they are none of that.

Andrej Karpathy goes deep in this 3 and half hour video on what LLMs are, how are they trained, programmed to fix their own biases and more. If you ever used ChatGPT or any other LLM, I think it’s a great watch to set your mental model straight and understand how to properly interact with these models and understand why they give good and bad results. It’s absolutely worth watching, it opened my eyes.

Blogposts, discussions, events

Livestream Recap: Migrating from Cypress to Playwright using AI

We had an amazing time with Jonathan on our recent livestream. We covered some great insights on testing, automation, and AI. Definitely worth checking out!

​Watch the replay on LinkedIn →​

99 Cypress Tips – Have You Joined Yet?

If you’re looking to level up your Cypress game, my course 99 Cypress Tips is exactly what you need.

And now, Gleb Bahmutov has added some fantastic video content!​

I’m Heading to the USA – Let’s Meet!

That’s right! I’ll be traveling to the US soon and would love to connect with fellow testers, developers, and Cypress enthusiasts. If you’re around and want to meet up, let me know!

​Let’s connect on LinkedIn→

Test automation tip

Did you know you can format your cy.log() commands? They support markdown syntax so you can add links or make the text italic or bold. 

 1  cy.log('hello world')
 2  cy.log('**bold**')
 3  cy.log('*accent*')
 4  cy.log('[read the docs](https://docs.cypress.io)')

Meme of the week

Filip Hric

Teaching testers about development, and developers about testing

filip@filiphric.sk, Senec, Slovakia 90301
​Unsubscribe · Preferences​

Filip Hric

I teach testers about development and developers about testing

Sign up for weekly tips on testing, development, and everything related. Unsubscribe anytime you feel like you had enough 😊

Read more from Filip Hric

Hello Reader, If you’ve been reading this newsletter for a while, you know that quality engineering is the hill I’ll always choose to stand on. And this week, I get to share something personal that ties directly into that. I’m joining Qodo I’ve been following Qodo for almost a year now, and I’ve been getting more and more impressed every day. So I’m thrilled to share that I’m joining Qodo as a DevRel engineer. Qodo is an enterprise multi-agent platform for AI-driven code reviews. As AI...

6 days ago • 2 min read

Hey Reader, An interesting thought is popping up in conversations around AI agents: the environment around the thing matters more than the thing itself. Last week I read about Harness engineering and it felt very familiar. As if it was tapping into instincts I already had. If you’ve ever debugged a flaky test only to find the problem was in the setup, not the assertion, that instinct will feel natural. But at the same time it also feels like an unfamiliar territory. It borrows the same...

13 days ago • 3 min read

Hello Reader, AI can generate code, arguably with a pretty decent quality. That’s not news anymore. The question that’s been forming in my head all week is different: how do we decide what should go into production? Writing code is not the hard part (arguably, it never was). The hard part is making sure the right code ships and the wrong code doesn’t. And right now, that selection problem is becoming the defining challenge of AI-assisted development. Last week has definitely showed this....

21 days ago • 5 min read