AI stopped being a promise for the future and moved into your code editor today. But between the hype and the fear, the honest conversation is missing: which tools are actually worth it, which skills do you need to sharpen, and can you trust the code it generates? That's what we talked about with Francisco Erramuspe in this Howdy Session.
There are plenty of tools; good judgment for picking them is rarer
Every week a new AI tool appears promising to change your life. Francisco and Darío cut through the noise: it's not about using all of them, it's about understanding what problem each one solves in your workflow. The tool is the means; your ability to know when to use it is still the real differentiator.
Can you trust AI-generated code?
The short answer: it depends, and that's exactly why you can't take your hands off the wheel. Generated code can save you hours, but it can also slip in subtle bugs or decisions that don't fit your architecture. The conversation keeps coming back to one key point: AI speeds things up, but the responsibility for what reaches production is still yours.
The skills that actually matter right now
If AI is writing the routine code, what's left for you? Exactly the most valuable part: understanding the problem, designing the solution, reviewing with a critical eye, and communicating clearly with your team. No model replaces those skills, and they're precisely what we look for in Howdy developers.
Watch the full Howdy Session
Hit play to hear the full conversation, with tool examples and the real challenges of building with AI. And if you're a developer in LATAM looking to grow while working remotely with great teams, check out the opportunities at Howdy.
If you want to go deeper on the tooling side, here's a look at the AI tools engineers actually rely on in production.
