Aspiring Data Analyst. I turn messy public data into clear answers using Python, and I document every cleaning decision and caveat along the way. I'm open to data analyst roles.
Tools: Python · pandas · NumPy · matplotlib · seaborn · GeoPandas · Jupyter · Git
1.35M police records joined with ONS population and deprivation data. London's overall crime rate is ~65% higher, but its violent-crime rate is the same; the gap comes from theft and robbery in crowded areas.
|
634K US domestic flights (July 2024). One IT outage (CrowdStrike, 19 Jul) caused more cancellations than a hurricane, and one airline group took 62% of them.
|
~785K 2023 job postings. SQL appears in ~58% of Data Analyst postings, and Python and Tableau pay more than SQL.
|
28 programs I built while learning Python (games, tools, and API and file automation) plus 40+ practice problems.
|
- Question first. Each project starts from a question, not a dataset.
- Show the caveats. Missing data, outliers and source disagreements are explained, not quietly dropped.
- Reproducible. Every project links its sources, and its charts can be rebuilt from code.





