Skip to main content

Mentor in the AI Era

5 min read

Tech Lead

Your juniors will use AI. Your job: teach them when to trust it, when to doubt it, and what 'good' looks like. That's mentorship now.

Eng Manager

You're mentoring people who learn differently. AI changes the feedback loop. Adapt your mentorship — less 'how to code,' more 'how to think.'

Qa

Juniors in QA will use AI for test generation. Mentor them on: what to test, what edge cases matter, when to override the AI. Judgment is the skill.

Mentor in the AI Era

TL;DR

  • Juniors using AI need different mentorship. Less "here's how to write a loop" — more "here's when to trust AI, when to overrule, and what good looks like."
  • The skills that still matter: problem decomposition, review rigor, ownership. Teach those. AI handles the syntax.
  • Yuki mentored people who learned to code from scratch. Now she mentors people who learn with AI as default. Same goal: build judgment.

Your juniors will use Cursor. They'll paste errors into ChatGPT. They might not know when to stop using AI and think. That's your job to fix.

What to Teach

  • When to use AI: Unfamiliar territory, boilerplate, debugging. Good.
  • When to think first: Core patterns, architecture, anything security-adjacent. AI can help; you lead.
  • Review rigor: AI output is draft, not final. Teach the checklist: What could break? What's missing? What's wrong for our context?
  • Ownership: "I used AI for this" is not an excuse. You own the output. You ship it. You fix it when it breaks.

What to Watch For

  • Over-reliance: Junior who can't explain their code. Who can't do basic tasks without AI. That's a crutch. Intervene.
  • Under-use: Junior who refuses to touch AI "because it's cheating." They're leaving speed on the table. Encourage experimentation.
  • Copy-paste without understanding: The classic. "What does this do?" — if they can't explain it, they didn't learn it. Drill that.

Adapt Your Style

You might have learned by reading the docs, debugging for hours, asking seniors. They learn by asking AI, getting an answer in 30 seconds, and moving on. That's fine — if they also internalize. Check for understanding. Ask "why?" often.

Quick Check

Your junior uses Cursor for everything. They paste errors into ChatGPT. What should you teach them that's DIFFERENT from pre-AI mentorship?

You mentor juniors who learned from docs and debugged for hours. You teach syntax, patterns, 'how to code.' That was the job.

Click "AI-era mentorship" to see the difference →

Do This Next

  1. Have one conversation with a junior about AI use. "How do you use it? What works? What doesn't? When do you turn it off?" Listen. Adjust your mentoring.
  2. Create one "AI review checklist" for your team. What do you always verify in AI-generated code? Share it. Make it the standard.