Welcome to my GitHub! I’m a data enthusiast with a unique blend of commercial and technical experience—from managing global FMCG brands to building robust data and ML solutions. My journey is all about connecting business needs with the power of analytics and machine learning.
- After years in Marketing & Account Management for brands like L’Oreal and Kraft Heinz, I realized how much business success depends on data-driven decisions.
- I made a bold switch: diving deep into Python, machine learning, and cloud tools to bridge the gap between business and tech.
- I thrive on solving real-world problems—whether that means optimizing sales, predicting outcomes, or automating workflows.
🧠 FMCG Promotion Copy Re-use Recommender
AI-powered semantic search and analysis tool for FMCG promotional copy. Enables marketers and trade marketing teams to find, classify, and re-use high-performing headlines using LLMs, emotion analysis, and KPI filters. Modular Python project with Streamlit dashboard—showcasing practical data science and modern NLP for marketing.
- Semantic search to find similar promotional headlines using AI embeddings
- Emotion analysis to detect tone (excitement, urgency, trust, joy, etc.)
- Zero-shot classification for promo types (Price-off, Bundle, Value-add, Sampling, Contest)
- KPI filtering (sales lift, ROI)
- SKU copy re-use recommendations
- Interactive dashboard via Streamlit
- Modular, agent-compatible architecture for extensibility
🌸 Wangi Perfume Recommender
AI-powered app for personalized perfume recommendations based on real customer reviews and user scent preferences, tailored for the Indonesian market.
- Scraped and analyzed 1,000+ perfume products and reviews from femaledaily.com
- Used NLP and semantic search to match user queries with suitable perfumes
- Built a real-time recommendation engine and deployed it as a web app with Streamlit and HuggingFace
- Delivered actionable insights for both shoppers and brands on fragrance trends and preferences
🏙️ NYC Airbnb ETL & Analytics
Automated an end-to-end pipeline to analyze Airbnb listings and market trends in NYC.
- Designed and orchestrated data workflows using Python, PostgreSQL, Apache Airflow, Great Expectations, Elasticsearch, and Kibana
- Delivered actionable insights for market performance and trends
📷 Road Image Classification using Computer Vision (CNN)
Developed a deep learning model to classify road images as “Clean” or “Dirty.”
- Implemented robust image preprocessing and model deployment with TensorFlow, Keras, Streamlit, and HuggingFace
- Deployed a scalable solution for real-world urban cleanliness monitoring
🧑🏻🎓 ML Classification Student Placement Prediction
Created a machine learning solution to forecast MBA graduate placement.
- Built and evaluated end-to-end ML pipelines with scikit-learn, XGBoost, and Streamlit
- Presented insights and results to stakeholders for data-driven decision-making
🛍️ Zara In-Store Sales Analysis
Analyzed the impact of product visibility on sales using descriptive and inferential statistics.
- Transformed raw retail data into strategic recommendations using pandas, matplotlib, and Tableau
- Provided clear, data-backed guidance for optimizing merchandising strategies
- How to translate messy, real-world data into strategies that drive results
- The art of storytelling with data—making complex insights accessible and actionable
- The value of persistence: every challenge (from data cleaning to deployment bugs) has made me a stronger, more adaptable problem solver
I love collaborating with curious minds and tackling new challenges. If you’re passionate about data, business, or tech—or you just want to chat about ideas—reach out!
Let’s build something impactful together!