Data Analyst

Aayush Damani

I'm finishing an MSc in Business Analytics & AI at Imperial College London, after a couple of years working in data and business analytics roles. This site has a few projects I'm proud of, with the full code and reports attached.

View Projects Get in touch

Featured work

See all 8 projects →

DescriptiveText analytics across 61,521 listings in 16 cities

Do Listings Lie? How Airbnb Language Predicts Price

Premium words like “luxury” and “stunning” predict nightly price three times more strongly than generic positive sentiment.

DescriptiveSQL analysis of 193,468 deliveries across 816 IPL matches, 2008-2020

Do the Numbers Back the Legends? Thirteen Years of IPL in SQL

AB de Villiers leads the strike-rate table among the tournament's highest-volume scorers, leads it again under the highest-pressure overs of an innings, and has more Player of the Match awards than anyone in IPL history. That holds across 13 years of data, not one standout season.

PredictiveK-Prototypes clustering + segment-specific regression on 5,329 e-commerce customers

Customer Segmentation & Heterogeneous Churn Dynamics

In most customer segments, ordering more predicts a higher chance of churning. In the highest-risk segment, it's the opposite: ordering more predicts they'll stay.

Applied AIBuilding an AI assistant on the official FIFA 2026 rulebook, then measuring where it still gets things wrong

Confidently Wrong: Testing a World Cup Rules Chatbot

The version that answered the most questions correctly was also the one that invented an answer to every single question it should have refused.

Ask about my work

I built this so you can ask about my background instead of reading six project write-ups. It reads this site and my CV and nothing else, it names the project or role each answer came from, and it tells you when something isn't covered rather than guessing. That last part is the point: my World Cup chatbot project measured how confidently these systems invent answers, so it would have been a poor look to ship one that does it.

Built with Claude Sonnet 5 and no retrieval database. The whole site comes to around 4,000 words, short enough to hand the model in one piece, so there is no search step for the right passage to get missed by.

Answers are written by an AI model reading only this site and my CV. It can still get things wrong, so check anything that matters against the CV or just ask me.