From learning to data-driven decisions.

Tools change. The loop doesn't.

Every project follows the same loop.

I'm learning the tools. Not just the theory.

1. DoneFoundationIntermediate · GSDC
2. DoneWork1 year of experience
3. NowData analyticsPython · pandas · SQL
4. NextData scienceStatistics · ML basics
5. NextAIModels and tools
6. NextData engineeringPipelines · ETL

A toolkit, with honest levels.

Confident means I've used it in a finished project. Building means I'm using it regularly. Exploring means I'm learning it now.

Learning in public.

Each one starts with a dataset and a question, built hands-on in Python, SQL, and pandas. Flip the switch to read the full story.

• Py

  01 · Python

  Apple App Store Data Analysis

  Which app genres dominate the App Store, and does price change that?

  Started with the raw App Store CSV, using only Python's csv module. Built a frequency table of app genres, then labelled every app free, cheap, affordable, or expensive by price to see whether genre and pricing move together.

  Tools: Python, CSV module, dictionaries, conditional logic

• SQL

  02 · SQL

  SQL Data Analysis with SQLite

  What can relational data tell you once you can actually query it?

  Connected to two SQLite databases through Python: a country-level factbook and a jobs.db dataset of college majors and job outcomes. Used sqlite3 with pandas to filter and join with SQL and pull results straight into dataframes.

  Tools: SQL, SQLite, pandas

• Py

  03 · Python

  Retail Sales Data Analysis

  Where is revenue actually coming from: city, category, or payment type?

  Analyzed retail orders from Pakistani cities, including Karachi, Lahore, and Islamabad, with product category, payment method, quantity, and revenue. Broke totals down by city, category, and payment method to see which combinations drove the most sales.

  Tools: Python, CSV module

• pd

  04 · Pandas

  Data Cleaning & Aggregation

  Real data is never clean. How much can you trust it before you fix it?

  Took messy real-world data, including laptop listings and a company dataset, and made it usable in pandas: fixed inconsistent text, handled missing values, corrected data types, and used groupby to summarize trends.

  Tools: Python, pandas

• pd

  05 · Pandas

  Movies Dataset Analysis

  Two datasets, one story. What happens when you merge them?

  Merged a 4,800+ movie dataset with a credits dataset, then parsed the nested JSON columns (genres, cast, crew) into real lists so they could be filtered and counted. Analyzed budget, revenue, and ratings together.

  Tools: Python, pandas, JSON

Data in motion.

A small live chart. Swap in your own Power BI or analysis data later.

Where I've been.

Let's turn data into decisions.

Open to data analyst, data engineer, and Power BI opportunities, internships included.