I'm a past Microsoft and Revolut intern pursuing a dual degree in Data Science & Engineering and Mathematics at UPC Barcelona. Passionate about AI, machine learning, and innovative technology solutions. I enjoy tackling complex problems and creating applications that make a positive impact.
- Double Bachelor's Degree in Data Science & Engineering, and in Mathematics
Universitat Politècnica de Catalunya (UPC Barcelona)
Expected Graduation: June 2027
-
guAIta - HackUPC 2026
Edge AI monitoring proof of concept for Collserola wild boar detection and operational response. Replaces costly manual surveillance with targeted, event-driven responses using local on-device inference.- Technologies: Node.js, Fastify, React, Arduino UNO Q, Edge Impulse, SQLite, Socket.IO
- Key Achievement: 🏆 3rd Place Winner - Qualcomm Challenge
-
SoLatAm - UAB THE HACK Hackathon 2025
AI-Powered Renewable Energy Transition Platform for Latin American communities.- Technologies: Python, NASA POWER API, OpenAI, Computer Vision, Streamlit, Folium
- Role: Led the data processing pipeline and solar potential mapping features
-
Design Decoder - Datathon FME 2024
AI tool to automate garment attribute recognition from images, freeing designers to focus on creative work.- Technologies: PyTorch, CLIP, XGBoost, OpenCV, Streamlit
- Key Achievement: 🏆 1st Place Winner at Datathon FME 2024
-
Street Style Decoder - HackUPC 2024
App that identifies clothing items from photos and recommends similar products from retail catalogs.- Technologies: CLIP, U-NET Semantic Segmentation, PyTorch, Cosine Similarity
- Role: Developed the semantic segmentation component to isolate clothing items
-
LightChaser - 2024 France Tech Arena
Network optimization solution using modified Dijkstra's algorithm to replan service paths after fiber failures.- Technologies: Graph Algorithms, Wavelength Optimization, Python
- Key Challenge: Optimizing wavelength allocation while maximizing service continuity
-
Chip Power Delivery Optimization - Datathon FME 2022
Solution to a variation of the Traveling Salesman Problem using Genetic Algorithms and Simulated Annealing.- Technologies: Genetic Algorithms, Simulated Annealing, Python
- Role: Algorithm design and implementation
-
Song Popularity Predictor
Machine learning project predicting Taylor Swift's song popularity using classical ML models.- Technologies: Python, Scikit-learn, Pandas, Matplotlib, Seaborn
- Models: Linear models, decision trees, ensemble methods, support vector machines
- Process: Data preprocessing, feature engineering, model training and evaluation
-
SIAM Student Programming Competition
Automated Market Maker (AMM) simulation with evolutionary optimization for liquidity provision strategies.- Technologies: Python, NumPy, Pandas, Evolutionary Algorithms
- Key Features: Conditional Value at Risk (CVaR) optimization, market dynamics simulation
-
Workgroup Manager
Cross-platform desktop application for coordinating student group projects and visualizing member availability.- Technologies: Python, tkinter, MySQL
- Key Features: Visual calendar for group availability, user accounts with authentication, group administration tools
-
Cinebus
Desktop app for finding movie listings with optimized public transport routes to cinemas in Barcelona.- Technologies: Python, BeautifulSoup, NetworkX, tkinter
- Key Features: Real-time movie listings via web scraping, integrated bus network navigation, interactive city map visualization
- Implementation: Combined city road graph with public transport data to calculate optimal routes
-
Fantasy Football
Algorithm-driven application for creating optimal fantasy football teams based on player statistics.- Technologies: C++11
- Key Features: Multiple team selection algorithms (exhaustive search, greedy, genetic algorithm)
- Optimization: Smart filtering of player database to reduce search space and improve performance
- Rust
I'm always open to new opportunities and collaborations. Let's connect!
