Stop the verification of engineers, as if it was 2021-AI’s workforce

You are already behind, if you are still checking engineers, as if it were 2021. We live because of what I think will be the most transformant technological change of our lives, even greater than the Internet.

The AI revolution accelerates at the rate at which most of us can’t even understand. This is not a noise. This is re -calibration of what building, creation and work means. The founders who are preparing now will lead in what will happen next. Those who will not be overtaken by five-person native startups that work at 10 times speed and precision.

So how do you hire programmers in the acceleration era?

You do not check them for how well the code write. You accuse them of seeing how well they organize it. Let me explain.

AI liquidity is actually a recent reading and writing skill

Each founder wants “AI programmer”. But this term can mean many things. Are you looking for someone who is building huge language models (LLM) in Python? Or maybe someone qualified using AI tools to escalate speed and reduce errors?

Most companies need a second. But they don’t always know how to ask for it. That is why AI or how well the programmer can move and operate a wide range of AI tools, becomes as critical as knowledge of a specific language or frame.

The instrument will change. But finishing learning how to operate recent AI assistants, assess their production and include it in your workflow? This is a lasting benefit.

What is AI-Orchestraver and why do you need it?

The AI orchestrator is today’s necessary archetype. They do not write manually every line of code-monitor, criticize, debug and reimbursement resulting from AI. They understand when to delegate to machines and when to operate your judgment. And they know how to communicate with AI agents such as colleagues.

At the same time, when AI is speedy, it’s not always right. And he certainly does not know the specific needs of your company. So the features you want to determine the priorities in employment are:

• Architecture – High levels can be enlarged and designing systems.
• Critical thinking -a price of compromises, making good decisions and choosing the right tools for work.
• Communication – It’s huge. How well can you explain your thinking to the robot? Ai does not make heuristics. You won’t get what you need if you can’t express what you want.

Just as we have not stopped teaching mathematics, because there are calculators, we cannot abandon basic programming skills just because AI writes the code. We need programmers who understand architecture, know when to trust artificial intelligence and know when to enter and fix what has been broken.

4 Ways to assess the competences of the AI engineer

In response to the spread of tools AI, my company was looking at the way we check technical talents. The time-honored technical interview process, challenges algorithm and coding tests specific to the language simply do not limit it anymore.

Here’s what to do instead:

• Simulate real problem solving. Ask candidates to build a function or debug the problem, but don’t let them write any code yourself. Instead, require them to operate tools such as ChatgPT or Claude, dividing the screen all the time so that you can watch how they interact with AI.
  
• Rate the hint. You are not looking for the right answer. You want to see how candidates develop a problem, monitor artificial intelligence and improve and intensify at his exit. This exercise is more about determining the clarity of thinking and communication of the candidate to master the syntax.
  
• Verify authenticity. Yes, people will try to cheat by sharing screens with someone else, having to impersonate them or resort to deep wardrobes. Therefore, you will want to insist on sharing a full screen and turning on the camera. Inform programmers that you are not trying to get “Gotcha” on them; You want to understand how they work with AI for the day.
  
• Test sentence. It is simple to get a working code from AI. A more complex skill is to know if it is a good code, it suits the system architecture and is the right solution to the problem. In all these steps you will want to check if they can spotless a critical thinking bar over a elementary copy effort.

What to remember about AI adoption

So we have built a training for each level of experience. For juniors, it is about slowing them down, helping them see where Ai controls them badly. For seniors, it is about education with AI integration without losing control. In both cases, the goal is to unlock real performance without prejudice to quality.

Accept this change, it creates the possibility

Yes, this transition to AI is terrifying. And yes, there will be turbulence. There will be works that disappear and recent ones that grow. But those who learn to screen, train and build bands around talent from the AI-Exeda function will write the future.

If you still hire engineers for what they can do yourself, you lack the heart. Start employing them based on how well they work with machines.

The future is not artificial intelligence compared to people. This is AI With People and those who adapt the fastest are winning.

Jacqueline Samira is the founder and general director Howdy.comwhich builds and manages the elite software engineering teams in Latin America.