My Projects

Supply Chain Data Processing System

Developed a system to efficiently process high-volume supply chain data, focusing on streaming order data. Created real-time KPI dashboards and demand...

Data Engineering Best Practices

This project implements an industrial-level data infrastructure solution for a city traffic department to efficiently collect, store, and analyse traffic data captured by swarm UAVs (drones) and static roadside cameras....

Knowledge Graph - Tourism and Waste Management

Built a knowledge graph providing insights into the relationship between tourism and waste management in Trento, Italy. Created system supporting...

Text-to-Speech Data Collection Pipeline

Designed and implemented a pipeline for speech data collection to train and deploy language models. Created system to make AI assistants more user-friendly by processing audio input from various sources..

Route Optimization and Recommendation System: Data Mining Project

The project addresses discrepancies in logistics, focusing on drivers deviating from planned routes and load specifications. Its goal is to minimize these deviations by recommending optimal standard routes based on drivers’ actual behavior and preferences. This includes generating company-wide standard routes, ...

Dynamic Creative Optimization

This project focuses on leveraging deep learning-based computer vision techniques to optimise creative content in mobile advertising. The project covers end-to-end automation—from data pipeline development to model deployment and monitoring.

Causal Inference Modelling

This project implements causal inference on real business data, to understand demand distribution and make business decisions.  Causal inference is the process of determining the independent, actual effect of a particular phenomenon that is a component of a larger system. 

Sales Prediction - Rossman Pharmaceuticals

This project Applies machine learning to forecast sales in pharmaceutical retail based on a real-world business problem from Kaggle. Created models to estimate..

Intelligent Job Recommendation System

This project develops an intelligent job recommendation system using large language model (LLM) to extract key entities from job descriptions, addressing the limitations of traditional keyword-based matching systems.

Fine-Grained Image Classification

This project aims to evaluate and enhance two state-of-the-art deep learning models to classify fine-grained images without part-based annotations, using only raw images and labels of benchmark and evaluation datasets.

High Performance Computing

Advanced Social Network Analysis