07. Hiring in the AI Era: Who to Hire and Who Not to Hire?

Headcount Anxiety: Budget Cuts and Talent Mismatch

Recently, I’ve heard a lot of complaints about hiring:

• “The budget was cut, but business targets went up. I don’t know who to hire.”
• “Experienced engineers on the market are too expensive, and I dare not hire junior engineers for fear of negative output.”
• “Technology iterates too fast. Will the expert hired today be eliminated by AI tomorrow?”

Behind these anxieties is the huge challenge traditional hiring logic faces in the AI era. We used to hire based on “Tech Labels”: hire a “Frontend familiar with Vue”, hire a “Backend with 5 years of Java experience”. But now, the value of these labels is rapidly being eroded by AI.

In the fifth article, we mentioned that AI will replace “Low-Judgment Roles”. So, the focus of hiring should shift to those “High-Judgment Roles”.

Hire “Delegatable People”, Not “Tech Labels”

In the AI era, the core principle of hiring is: Hire people who can independently complete a full business loop and be responsible for the final result.

Such talents often have the following characteristics:

1. Strong Sense of Ownership: They don’t limit themselves to a specific tech stack or functional boundary. When seeing a problem, the first reaction is “how to solve it”, not “this is not my job”.
2. Excellent Judgment: They can understand the business context and make the optimal choice among multiple solutions provided by AI based on the actual situation. They can identify “High-Quality Bugs” generated by AI and correct them.
3. Strong Learning and Adaptability: They are not obsessed with a specific technology but focus on solving problems. After AI appeared, they quickly turned AI into their “External Brain” instead of being eliminated by AI.

These characteristics collectively form a term: “Delegatable People”. They are the kind of people to whom you can throw a complex problem, and they can use all existing tools (including AI) to dismantle and solve it, and finally deliver an operable solution.

Senior Engineers: From “Too Expensive” to “Super Value”

When budgets are tight, many managers tend to hire “a few more junior engineers”, thinking it’s more cost-effective. In the AI era, this might be a huge misjudgment.

Let’s review the conclusion of the second article: In the AI era, the advantage of small teams is not efficiency, but judgment density.

A senior engineer might be able to use AI to achieve the output of 3-5 junior engineers in the past, but his core value lies in his “Judgment” and “Architectural Ability”. He can:

• Identify potential architectural hazards in AI-generated code.
• Weigh trade-offs among multiple technical solutions.
• Provide stable technical direction to prevent the team from getting lost in the “Linear Acceleration Illusion” created by AI.

The errors of senior engineers are local and can be discovered and corrected in time through Code Review. The low-judgment group, empowered by AI, makes systemic errors. Once they erupt, they are often irreversible.

Therefore, in the AI era, senior engineers are not “too expensive”, but super value.

(Of course, the premise here is: this senior engineer still maintains a sense of responsibility for the real system and can personally get down to solving the thorniest problems, rather than just being a “PPT Architect” who stays at the solution and review level.)

New Job Title Suggestions: Product Engineer / System Engineer

Traditional job titles (Frontend, Backend, QA, Ops) can no longer accurately describe the talent we truly need.

I suggest considering the following job titles:

• Product Engineer: Focuses on the end-to-end delivery of a product functional domain. They understand the business and can use AI to quickly implement the entire link from UI to data flow.
• System Engineer: Focuses on architectural design, performance optimization, and stability assurance of a complex technical domain. They can gain insight into the risks brought by AI and provide safe and reliable engineering practices.

These positions emphasize “Ownership” and “Responsibility” more than just “Tech Stack”.

Of course, the job title is not the goal, but a tool to align expectations. You can still call them “Frontend Engineers”, but you need to tell them clearly: your output is “Functionality”, not “Pages”.

Interview Strategy: Assessing “AI Driving Ability” and “Judgment”

Since hiring standards have changed, interview strategies must change too.

• The Significance of Whiteboard Coding Drops: Asking an interviewee to hand-write a bubble sort can no longer effectively evaluate whether they can solve real problems in the AI era.
• Scenario-based Questions are King:
  • “Given such a business requirement, how would you break it down? Which tech stacks would you use? If you use AI to assist, what would you let AI do, and what would you focus on checking?”
  • “Describe a complex problem you recently solved with AI. Where did AI help you? Where did you correct AI’s output?”
  • “If AI generated a piece of seemingly perfect asynchronous processing code for you, but you found it causes deadlocks under certain extreme conditions, how would you troubleshoot and solve it?”

These questions aim to assess the interviewee’s business understanding, system design ability, critical thinking, and AI driving ability.

Conclusion: HC Should Be Spent on “Brains”, Not “Hands”

AI is rapidly increasing the output of our “hands”, which means Headcount should be invested more in “Brains”.

Hire the right person, and they can use AI to leverage the efficiency of the entire team and avoid falling into AI traps. Hire the wrong person, and they might bring AI to create ten times the problems at ten times the speed.

If an organization has no red line for judgment, then more HC only amplifies risk.

In the AI era, hiring is no longer about filling technical gaps, but investing in “Decision Power” and “Responsibility”.