DA_Python

End-to-end data analysis project exploring layoffs in technology companies in the context of AI adoption.
Focus: real-world data pipeline, statistical validation, and business-oriented insights.

Live dashboard

Blog

---

What This Project Demonstrates

This project is not just a set of notebooks. It demonstrates the ability to:

• Work with real-world, messy datasets
• Combine multiple data sources (structured + text-based)
• Design a data processing pipeline
• Apply statistical thinking to business problems
• Deliver results via a clear, interactive dashboard

---

Key Capabilities

• Data acquisition & enrichment

  • integration of external data sources
  • text enrichment via APIs

• Data processing

  • cleaning, normalization, and transformation
  • handling incomplete and inconsistent data

• Feature engineering

  • categorization of layoff reasons
  • structuring unstructured text into analyzable formats

• Analysis

  • hypothesis testing
  • trend identification across companies and time

• Visualization

  • building a Tableau dashboard for insight communication

---

Tech Stack

• Python (core analysis)
• Jupyter Notebook (exploration & iteration)
• pandas, SciPy (data processing & statistics)
• SQL (data validation and structuring – workflow layer)
• Apify (data acquisition – article context)
• OpenAI API (text enrichment – article context)
• Tableau (final dashboard)
• Figma (dashboard design – article context)
• Google Maps Platform API (geolocation enrichment – article context)

---

Analytical Workflow

1. Data Acquisition

   • collecting layoff datasets
   • enriching records with external sources and articles

2. Data Preparation

   • cleaning and standardizing data
   • merging multiple datasets

3. Feature Engineering

   • extracting and categorizing layoff reasons
   • transforming text data into structured variables

4. Analysis

   • statistical hypothesis testing
   • identifying patterns and trends

5. Visualization

   • translating analysis into a Tableau dashboard
   • enabling exploration of insights

---

Key Insight Focus

The project investigates questions such as:

• Are layoffs increasing in correlation with AI adoption?
• What are the dominant reasons companies report for layoffs?
• How do layoffs vary across industries and time periods?

→ The answers are presented in the dashboard, not hardcoded in the repository.

---

Repository Structure

The repository reflects an exploratory analytics workflow:

• data_layoffs.ipynb – core dataset processing
• duvody_v_layooffs.ipynb – layoff reason analysis
• flitr_chybejicich_clanku.ipynb – filtering missing article data
• kategorizace_duvodu_propusteni_15_05.ipynb – categorization logic
• tabulka_pro_testovani.ipynb – dataset preparation for testing
• zaloha_data_layoffs.ipynb – backup work
• pocte_url_v_datasetu_nedele.py – helper script
• data layoffs/ – dataset storage

---

Quick Start

git clone https://github.com/VerunkaZ/DA_Python.git
cd DA_Python
pip install jupyter pandas scipy
jupyter notebook