As a student of Artificial Intelligence at the Universitat Politècnica de Catalunya, I have a strong interest in applying machine learning and AI to address practical challenges. My curiosity motivates me to learn more about the field of AI, and I appreciate opportunities to test my skills in settings like hackathons. Through my studies and various projects, I have developed a solid foundation in Python, which is my strongest programming language. I also have experience with other languages such as R, SQL, C++, and C. This background supports my efforts to develop practical solutions within the realm of technology.

• Catalan: Native
• Spanish: Native
• English: Advanced (C1 Cambridge Advanced)

• HackUPC Champion: Led my team to victory at the HackUPC challenge, where we developed an AI-based solution to analyze transaction data, showcasing our prowess in machine learning and data analysis.
• Challenge Auditoria Winner: With other two colleagues we created a project capable of predicting the inventory impairment using advanced AI models such as autoencoders (NN), advanced time series algorithms, EBM for explaining the results, word embeddings and even clustering where we emerged to create a model user friendly in order to solve this real problem for auditors.
• Recognized with a Governmental Distinction for achieving over 9.5/10 in the University Entry Exam scores.
• Acted as a mentor for newcomers at Universitat Politècnica de Catalunya, offering guidance and support in their academic and professional pursuits in AI.

• Programming Languages: Proficient in Python, R, SQL, C++, C, and PDDL.
• Machine Learning & AI: Extensive knowledge in neural networks, deep learning, AI algorithms, and natural language processing.
• Data Analysis & Statistics: Expertise in leveraging data analysis, mathematics, and statistics for predictive modeling and actionable insights.
• Parallelization: Experienced in enhancing algorithm performance through effective parallelization techniques.

• HackUPC Champion & Challenge Auditoria Winner
• Cirrhosis Survival Prediction: Developed ML models to accurately predict cirrhosis patient survival rates, achieving an F1-score > 0.8.
• Bicycle Renting Service AI Optimization (Bicing): Optimized bicycle redistribution logistics using local search algorithms like Simulated Annealing to reduce operational costs.
• Statistical Study of Spotify’s Top Songs: Analyzed Spotify’s top tracks using ML and statistical methods to identify improvement areas for the company.
• Research on Automatic Systems: Investigated the use of Arduino for creating an efficient automatic watering system, reducing water consumption through smart scheduling.

• Email: [email protected]
• LinkedIN: Roger Baiges Trilla