Build data projects
that get you interviews.
Not a tutorial you follow along with. Real briefs, real datasets, and me alongside you every week: live office hours and answers in Discord. Every finished project publishes straight to your portfolio.
Pasquale Bruno
I've spent 10+ years building analytics at Google and Meta, and sat on the other side of the table for hundreds of data analyst and data scientist interviews.
I write every brief, record every lesson, and answer your questions myself.
Read bio →SUM(payment_value) AS revenue
FROM olist.orders GROUP BY m
Hand-in-hand support, not a self-serve library.
Weekly office hours and real answers in Discord, from one subscription that includes every project.
Follow whichever fits how you learn, at your own pace.
SQL, Python, Streamlit: what analysts use on the job.
The same way analysts at top tech companies use LLMs.
Ask questions, get real answers from a Google and Meta data scientist, not a bot.
Bring what's blocking you straight to me, live, every week.
Next session: Thursday, Oct 15 · 5pm UTC
A monthly 30-min call, just you and me. Walk through your project, get unstuck on something specific, or talk through your career transition.
I personally review your finished portfolio and LinkedIn profile, and leave feedback before you send them to recruiters.
Live now
Each one is a business brief with a deadline, not a lesson plan.
A marketplace health analysis
You ship: a marketplace health writeup
Monday morning, week one. Your manager hands you the BigQuery schema for a Brazilian e-commerce marketplace and wants a picture of the business (revenue, customers, sellers, delivery, satisfaction) before Friday's exec review.
with Pasquale · ex-Google, ex-Meta
A revenue performance analysis
You ship: a 6-slide board story
The CMO wants to understand what drove last year's best and worst months. You'll run an end-to-end EDA on a real UK retail dataset, find the drivers, and turn the analysis into a 6-slide story for the board.
with Pasquale · ex-Google, ex-Meta
A sales funnel diagnosis
You ship: a Looker Studio dashboard
You're the analyst at a UK fashion e-commerce brand. Revenue is missing target and the team needs a diagnosis. You'll build the funnel end-to-end in SQL, find the real drop-offs, and ship a Looker Studio dashboard plus a stakeholder recommendation.
with Pasquale · ex-Google, ex-Meta
A checkout experiment analysis
You ship: a Streamlit decision tool
The checkout experiment finished Friday. Conversion is up, but is it safe to ship? You'll run a real A/B test analysis on a 14-day, 10k-session experiment, segment by device, and ship a Streamlit tool the PM can use without opening a notebook.
with Pasquale · ex-Google, ex-Meta
A user feedback triage
You ship: a 4-slide recommendation deck
Roadmap lock is Thursday and there are 3 engineers, not 30. Everyone's got a theory: ads, buffering, Shorts. You'll classify 2,170 real App Store reviews with the Claude API, rank the complaints, filter to what's actually fixable, and ship a 4-slide recommendation deck.
with Pasquale · ex-Google, ex-Meta
Churn Prediction, House Price Prediction, Customer Segmentation, Fraud Detection, Ticket Routing
Coming Nov–Dec 2026
See what's coming →
I've interviewed hundreds of data candidates at Google and Meta. I know what makes a hiring manager keep reading.
I design and review every guided project myself, drawing on briefs and datasets modelled on the real work of a product and business analyst at a large tech company.
If you get stuck, you're emailing me, not a support queue.
Why these are different
Most project-based courses stop at a notebook. Guided Projects end with something you'd actually put in front of a hiring manager.
Releasing soon
Coming Nov–Dec 2026, included in your subscription the day they ship.
Churn Prediction
Telco's retention team wants a list of who's about to churn, but a simple rule beats a black-box model if nobody trusts it. You'll build a binary classifier on real Telco data, handle class imbalance, and rank the 50 riskiest customers, only to find the highest-risk customer often isn't the most valuable one to save.
House Price Prediction
A regression model on the Ames Housing dataset should nail house prices: square footage, bedrooms, done. Except two identical houses, same size, can be worth double depending on the neighborhood. You'll handle outliers, engineer features, and find out what the model actually leans on.
Customer Segmentation
Online Retail wants to know which customer segments actually matter. You'll run RFM analysis and K-means clustering on real transaction data, and find that the biggest segment by total revenue isn't the most profitable one per customer.
Fraud Detection
A fraud model scoring 99.9% accuracy sounds great, until you realize it's catching almost none of the actual fraud. You'll work a real, heavily imbalanced credit card dataset, apply SMOTE, and learn why precision, recall, and F1 are the only metrics that matter here.
Ticket Routing
Support tickets should be easy to auto-route, until you check where the model actually gets it wrong. You'll build a TF-IDF + classifier pipeline on real CFPB consumer complaints, and find it's not the common categories that break it, it's the rare-but-critical ones.
Tell us the business problem you want to practice. We're building the next batch from requests.
Send us a brief →Every project, weekly office hours, and direct access to me.
Free
$0/forever
For anyone with projects ready to share.
- ✓ Portfolio website with unlimited projects
- ✓ Personalised URL
- ✓ No design nor technical skills needed
- ✓ All your education and experience info in one place
Job Ready
$99/ month
For people investing in a data career.
- ✓ Portfolio website builder PRO
- ✓ All current + future Guided Projects
- ✓ Discord support channel for any questions, answered by me
- ✓ Weekly live office hours with me
- ✓ 1:1 monthly call with me (first 20 members)
- ✓ Async portfolio & LinkedIn reviews (first 20 members)
- ✓ Cancel anytime
Cancel anytime. Your portfolio and published projects stay yours either way.
Not sure if it's right for you?
Book a free 15-minute call with me. We'll talk through where you are and whether a Guided Project makes sense.
Everything else
How much time do I need each week?+
Projects run 5–10 hours total and are split into chapters, so you can do one sitting at a time. Most people finish a project across two or three weeks alongside a job.
Do I need to know SQL or Python already?+
The beginner projects assume you can write a basic SELECT and open a notebook. If you're past that, start with Olist or EDA & Storytelling. The intermediate ones expect you to be comfortable joining tables and working in pandas.
What exactly ends up on my portfolio?+
A written case study with your own results: the brief, your approach, your code and charts, and your recommendation. It gets a public URL on your portfolio and is indexed by search engines.
Can I cancel, and what happens to my work?+
Cancel anytime. Anything you've already published stays on your portfolio - you just lose access to unfinished and future projects.
Do I need my own Claude API key?+
The projects that use the Claude API walk you through getting a key. Usage for a full project costs well under a dollar at current rates.
Start with one project.
5 projects are live and 5 more are on the way. Each takes 5–10 hours, and when you finish, it goes on your portfolio. If you get stuck, ask me.