Why Good Data Scientists Fail Interviews

Good morning Data Scientists!

Fed up of getting interview, only to receive the dreaded “Unfortunately…” email?

(Or worse — not hearing back at all.)

Last week, my client James got an offer from a London-based dating app.

What’s interesting is that nothing about James was “perfect”.

He didn’t memorise 100s of STAR answers.
He didn’t try to sound clever.
And he definitely didn’t treat interviews like exams.

Instead, we focused on how he showed up in each round.

Here’s how we approached it 👇

Screening interview
The goal wasn’t to impress.

It was to:

• Clearly summarise his experience

• Explain why he was interested in the role

• Keep the tone conversational

Most people over-answer here.

James focused on clarity and motivation.

Technical interview
This is where most Data Scientists go wrong.

Instead of walking line-by-line through a notebook, James:

• Summarised his overall approach

• Explained the results first

• Talked through why he made certain choices

Interviewers don’t want a tutorial.

They want to know how you think.

Final round
We made a mindset shift here.

This wasn’t a test to pass.
It was a conversation with future colleagues.

So James focused on:

• Higher-level product and data strategy

• How he’d approach problems in their context

• Asking thoughtful questions about impact and priorities

That’s what separated him from other “technically strong” candidates.

The big takeaway?

Getting offers isn’t about having the strongest technical skills.

It’s about:

• Communicating clearly

• Showing good judgment

• And making interviewers think:
  “I’d enjoy working with this person.”

If you’re consistently landing interviews but not getting offers,
it’s a communication problem, not a skills problem.

If this resonates, just hit reply and tell me which interview stage you’re getting stuck at.

God Bless,

Albert