Only have 7 days to prepare for a Technical interview? Here’s exactly how to do it.

Hey all, welcome back to my newsletter! Thank you for being here.

Today, I’m sharing a 7-day plan for prepping for Data Science interviews.

In an ideal world you would have weeks to prepare for an interview. But life doesn’t always work that way, sometimes you just have a week, or maybe even less.

But, the good news is: You can prep for and ace an interview within 7 days. You just have to lock in for that week.

If you want to become good at system design, join Neo's system design newsletter with 160K+ software engineers: newsletter.systemdesign.one

Day -7: Research the company

You need to know the company you’re interviewing for. I don’t care how small the company is, you MUST spend the time learning about them, including:

• Earnings calls

• Customer reviews

• Competitor research

• Product updates or launches

Pro tip: Use Research mode on Claude or Deep Research on ChatGPT. They can do extensive research across thousands of data sources, which will help you save HOURS as you’re prepping for these interviews.

Day -6: Learn company's metrics + start SQL practice

Today's focus is understanding how this company actually makes money and measures success.

Here's what you're going to do:

• Learn how company's metrics inform strategy

• Mock up fake data with Claude or ChatGPT

• Recreate metrics in SQL

Here’s my prompt:

"Based on the earnings calls, product launches, and tech blog posts you summarized for [Company Name]:

-What are the key metrics for this company? -What are 3-5 datasets the company would have? -Generate data for these datasets with 20-100 rows."

Now practice writing SQL queries against this fake data.

Trust me, this approach is way more effective than doing random LeetCode SQL problems about movie databases.

Day -5: Case study mock interviews

Case studies are make-or-break for DS interviews. I know this because I used to be an interviewer at Meta and Amazon… and this is the interview that people often failed.

These test your analytical thinking AND business sense — two things that separate great data scientists from bad ones.

Ask Claude / ChatGPT to generate 3 case study questions using the deep research it's done (in Step 1). Make sure these case studies cover:

• How to identify growth opportunities

• How to manage & communicate metric tradeoffs

• How to apply statistics to real business problems

Pro-tip: Turn on Extended thinking in Claude and have them answer the same questions. I find it very helpful to see & learn from their thinking process. You'll start to notice patterns in how to structure your responses.

Day -4: Practice for DS&A interviews

Time to simulate the real thing.

Have Claude simulate an interview with you. At the end, ask for an evaluation of your performance.

Again, turn on Extended Thinking Mode so you can learn how an interviewer evaluates you in an interview.

Day -3: Practice for SQL interviews

Today is about simulating a real SQL interview environment.

I recommend using Interview Master, which we built to simulate real SQL interviews and ace your next interview.

I cannot overstate this: SQL is the most important skill for a data professional. I've interviewed for Data Science roles at 20+ companies, and I had a SQL interview at almost every single one.

Day -2: More case study interviews

Yup, we're doing more case studies.

Why? Because Data Science roles especially are heavily focused on case study performance,

See Day -5 for the details, but generate new questions this time. You want to see different scenarios and practice adapting your framework.

Day -1: Polish behavioral interviews

Search the web to find the behaviors that make people successful at the company.

Look for things like:

• Leadership principles (if it's Amazon)

• Company values and culture

• Success stories from current employees

Now tailor your STAR stories to highlight these exact behaviors. You want to demonstrate that you are good cultural fit for the company.

You got this.