AI SKILLS BY ROLE

AI skills for data analysts

Data analysts use AI to write SQL, clean data and explain results. AI makes them faster, and it also makes confident mistakes easy: a join that double counts, a filter that silently drops rows, a trend that is really a definition change. The strongest analysts treat every AI query as a draft to verify.

How AI shows up in the work

Where data analysts use AI every day, and where it goes wrong.

01

Writing queries

Generating SQL and Python from a plain-English request. Joins and filters are where AI errors hide.

02

Explaining results

Turning a chart into a narrative for stakeholders. AI tends to state causes the data does not support.

03

Cleaning and checking

Handling nulls, outliers and duplicates. Strong analysts ask what the AI removed, not just what it kept.

What strong AI use looks like

The habits that separate a capable data analyst from one who copies the answer.

Interview questions that test AI skills

Ask for a specific example each time. Vague answers are a signal in themselves.

  1. 01
    AI wrote a query that returns revenue 18% higher than last month's report. What do you check first?What it reveals: Debugging instinct: grain, joins, duplicates, filters.
  2. 02
    How do you validate SQL you did not write yourself?What it reveals: A repeatable verification habit.
  3. 03
    An AI explanation says churn rose because of the price change. How would you test that?What it reveals: Causal reasoning and willingness to challenge a fluent story.
  4. 04
    Tell me about a time an AI tool gave you a wrong number. How did you catch it?What it reveals: Real experience with AI errors.
  5. 05
    When is it faster to write the query yourself?What it reveals: Judgment about where AI helps.
  6. 06
    A stakeholder pastes an AI chart into a deck. What do you need to know before you sign off?What it reveals: Ownership of numbers that leave the team.
  7. 07
    How do you keep metric definitions consistent when people generate their own queries with AI?What it reveals: Thinking about data governance.
  8. 08
    What data would you never paste into an AI tool?What it reveals: Privacy and security awareness.

A sample task with an AI mistake to catch

Questions show what candidates say. A task shows what they do when the AI is confidently wrong.

The situation

The head of sales wants to know why Q3 conversion fell. The AI colleague has queried the CRM and reports conversion dropped from 24% to 17%, and blames the new pricing page.

What the AI got wrong

The query joins opportunities to activities, so deals with many emails are counted several times. Deduplicated, conversion fell only to 22%, and the drop is concentrated in one region with a new sales team.

What a strong candidate does

Checks row counts against the CRM total, finds the fan-out join, reruns at the right grain and reports the regional pattern instead of the pricing story.

Red flags

Signs a candidate depends on AI rather than working with it.

Answerr measures this for you.

These tools are free. Our main work is measuring how well people work with AI, so employers can hire on evidence and teams can build the skill. Hiring? See AI interview questions by role.

01

AI capability

How well a person works with AI on real work, measured across Judgment, Reasoning, Critical thinking, Questioning and Adaptability.

What AI capability is
02

AI in hiring

Candidates work a real task from the role alongside Marcus, an AI colleague who is sometimes wrong. You see who checked, who questioned and who just accepted.

Try the hiring demo
03

AIQ™

A verifiable score for how you work with AI, with the evidence behind every rating. Teams use it to measure and build AI capability. Your first one is free.

How AIQ is measured

Questions, answered.

Using AI to draft queries, clean data and explain results while verifying joins, row counts and definitions, and separating what the data shows from the story the AI tells about it.

See how your data analyst candidates work with AI.

Each candidate works a task from your role alongside an AI colleague. You get an AIQ™ score with the evidence behind every rating.