Last newsletter I asked what you wanted to read next. It was a tie — data project vs Claude tutorial. I'm going with Claude this week and promise the data project is coming next.
For a long time I used Claude the same way most people use it. Open a tab, ask a question, or post your email/post/newsletter and ask Claude to check spelling or improve. Useful, but nothing life-changing.
Then I saw how my husband uses it.
He has separate projects for different areas of his life. A project for work (split into different areas of his work). A project for research. A project for personal stuff. And then — he built a gym programme app. From scratch. He hosted it on GitHub. And after than he built a news aggregator app that also shows stock prices. My husband is a finance person. Not a developer. Not a data person. Just someone who figured out how to use the tools available to him.
I was genuinely shocked. So I asked him a million questions to improve my Claude set up. Let me share what I’ve learnt.
Before we proceed, a small ad. As always, your clicks on ads are really appreciated as it covers hosting fees.
5 minutes. Every AI story that actually matters.
The AI Report distills the day's most important AI news into one free 5-minute read. No jargon, no filler — just what 400,000+ business leaders need to know before their first meeting.
Switching from ChatGPT? Here's how to bring your history across
A lot of people are moving from ChatGPT to Claude right now. Here's how to bring your conversation history with you.
Export from ChatGPT:
ChatGPT → profile icon → Settings
Data Controls → Export Data
You'll get an email with a download link (can take 24h)
Download and unzip the file
Inside you'll find a conversations.json file

Upload to Claude: Go to Settings → Capabilities → Under Memory section you can find Import memory from other AI provider → Start Import

Set your preferences
This is the first thing to do. Claude has a preferences section where you tell it how you want it to behave across every single conversation — permanently.
How to find it: open Claude → click your profile icon in the bottom left → Settings → Profile.

Here you tell Claude things like how you want responses formatted, your tone preference, your location, your job, anything relevant to how you work.
My preferences include: write in British English, don't use em dashes, avoid AI clichés, use bold for key insights not for headers. I also told it I work in data analytics, I create LinkedIn content, and I'm based in Abu Dhabi.
Now every conversation starts with Claude already knowing this. I never repeat myself.
Set up Projects
Projects are folders inside Claude where you store context that stays attached to every conversation in that project. This is what my husband showed me and what completely changed how I use it.
How to find it: left sidebar → New Project → give it a name (e.g. "LinkedIn Content", "Work Analysis", "Job Search")

You will see a field that says "What are you trying to achieve?". Don't leave it blank.
This is where you paste context Claude needs for that specific type of work. Think of it as a briefing document. The more specific you are, the better Claude performs inside that project.
Good things to include:
What this project is for and what you're trying to achieve
Who you are in the context of this project (your role, your experience, your audience if relevant)
Any rules or preferences — tone, format, things to avoid
Background context Claude would need to help you properly
But here's the thing — most people don't know what to write there or where to start. The easiest way is to ask Claude to interview you. Say: "I want to set up a project for my LinkedIn content. Please interview me — ask me questions one by one about who I am, my audience, my writing style, and my goals." Answer the questions, then ask Claude to turn the answers into a structured document. Ask Claude to save it as .md files and also give you a summary that you paste into ‘What are you trying to achieve?’
I have two files for my content project:
aboutme.md — who I am, my background, my audience, my content pillars, what I do and don't write about.
toneofvoice.md — specific rules about how I write. British English, no em dashes, no AI clichés, no headers in LinkedIn posts, bold for key insights only.
Save these as .md files and upload them into your project alongside the instructions. Claude reads everything at the start of every conversation in that project.
Instead of explaining every time that I write for aspiring data analysts, that I use British English, that I don't want headers — Claude already knows.
Also, you can edit these .md files as you go. For example, I don’t like a specific phrase Claude uses - then I put a prompt ‘update tone_of_voice.md file to never use the phrase X’

Let Claude remember things
Claude has a memory feature that carries information across conversations even outside of projects.
How to find it: Settings → Memory → make sure it's turned on.

When something important comes up — a preference, a fact about your work, something you mentioned — Claude can save it and use it in future conversations. You can view and edit your memories any time in settings.
It's not perfect, but it means Claude gets gradually more useful the more you use it.
The setup takes time but it’s worth it
Preferences, one project, memory turned on. That's it.
After that, every conversation you have with Claude is better than it would have been without it. And the more you use it, the more it learns how you work.
Keep pushing 💪,
Karina
Need more help?
Just starting with Python? Wondering if programming is for you?
Master key data analysis tasks like cleaning, filtering, pivot and grouping data using Pandas, and learn how to present your insights visually with Matplotlib with ‘Data Analysis with Python’ masterclass.
Building your portfolio?
Grab the Complete EDA Portfolio Project — a full e-commerce analysis (ShopTrend 2024) with Python notebook, realistic dataset, portfolio template, and step-by-step workflow. See exactly how to structure professional portfolio projects.
Grab your Pandas CheatSheet here. Everything you need to know about Pandas - from file operations to visualisations in one place.

Data Analyst & Data Scientist