I have been on academic leave from UPC since February 2020 to work at the two startups I co-founded: first Amalfi Analytics, then teler.ai.
This page remains partially active. I still update it now and then to reflect my reduced academic activity. The UPC email address will continue working; see the Contact tab. You can find more information about my current work on my LinkedIn profile.
I will not be able to take any interns or PhD students in the foreseeable future.
I used to work at the Department of Computer Science of Universitat Politècnica de Catalunya - BarcelonaTech (UPC).
My research is mostly on Machine Learning and Data Mining, both in their algorithmic aspects and their applications. I am particularly interested in the applications of Data Science to Social Good.
In 2017, I co-founded Amalfi Analytics. We created data analytics platforms that help with the management of complex healthcare organizations, using Machine Learning. We helped hospitals provide better care for their patients and better working conditions for their professionals. The ultimate goal was to keep healthcare sustainable, accessible, and fair. Amalfi was acquired by the Relyens group in December 2023 - though I remained as CEO.
After Amalfi came to an end in early 2026,
I co-founded teler.ai.
Teler.ai is a virtual data scientist that lets teams explore, model and understand their data using natural language.
It delivers reproducible, auditable analyses, helping domain experts get reliable answers without depending on overloaded data teams.
News:
- September 2026: Co-organized the 11th edition of SoGood, the Workshop on Data Science for Social Good. Held in conjunction with ECML PKDD 2026 in Naples (Italy), September 7th, 2026.
- June 2026: Published a compilation of the industrial projects I worked on while at UPC, with topics including datacenter management, healthcare and life sciences, social network analysis, sports analytics and others.
- April 2026: Co-founded teler.ai and started building the cross-sector platform for organizations that collect data and don't have the capacity to analyze it.
- September 2025: Anniversary edition of SoGood. Co-organized the 10th edition of SoGood, the Workshop on Data Science for Social Good. Associated to ECML PKDD 2025 in Porto (Portugal), September 9th, 2025.
- September 2025: Co-chair of the Nectar Track of ECML PKDD 2025, Porto (Portugal).
- September 2024: Co-organized the 9th edition of SoGood, the Workshop on Data Science for Social Good. Associated to ECML PKDD 2024 in Vilnius (Lithuania).
- August 2024: I gave the talk Machine Learning for Clinical Management at the Applied Data Science Track of KDD 2024, Barcelona, August 2024.
- June 7th, 2023: Gilles Blondel defended his PhD thesis "Causal discovery and prediction: methods and algorithms". It presents two advances in Pearl's causal graph / do-calculus formalism: One, incorporating prior knowledge into causal discovery algorithms, and another defining and inferring the effect of interventions in dynamical systems. Co-supervised with Marta Arias.
- April 2023: New paper in a top Data Science journal: A case study of improving a non-technical losses detection system through explainability. Data Mining and Knowledge Discovery (2023), Special Issue on Explainable and Interpretable Machine Learning and Data Mining.
- December 2022: New paper Interpretable prediction of mortality in liver transplant recipients based on machine learning, with Xiao Zhang and Jaume Baixeries, in Computers in Biology and Medicine.
- September 2022: Conference communication, Disease classification risk through machine learning algorithms: Lessons learned from COVID-19. Xiao Zhang, Maria Barros, Maria Paula Gómez, Entela Kondi, Fritz Diekmann, Chloë Ballesté, Marián Irazábal, P. Montagud Marrahi, E. Sánchez Álvarez, M. Blasco, Martí Manyalich, Ricard Gavaldà, Pedro Ventura Aguiar, Jaume Baixeries. 29th International Conference of the Transplantation Society. Buenos Aires, 11th-14th September 2022. Abstract
- January 18th, 2022: Javier Fernández de la Rosa defended his PhD thesis on using Machine Learning in soccer analytics. An industrial doctorate carried out at Futbol Club Barcelona. Luke Bornn was the main supervisor.
- September 2021: New paper out, Development and Validation of a Model to Predict Severe Hospital-Acquired Acute Kidney Injury in Non-Critically Ill Patients. Team led by Alfons Segarra. Journal of Clinical Medicine 2021, 10(17), 3959; doi:10.3390/jcm10173959.
- September 2021: We held the Sixth Workshop on Data Science for Social Good (SoGood 2021) on Sept 13th. 8 great papers and keynote talks by Paul Lukowicz, Fosca Gianotti, and Dino Pedreschi.
- April 2021: A paper on clustering patients with Idiopathic Pulmonary Fibrosis, with researchers at Bellvitge University Hospital (Barcelona), published in the ERJ Open Research Journal of the European Respiratory Society.
- October 2020: Survey paper on AI in critical care in the light of COVID-19: Artificial Intelligence for clinical decision support in critical care, required and accelerated by COVID-19. Miia Jansson, Juanjo Rubio, Ricard Gavaldà, Jordi Rello. Journal of Anaesthesia Critical Care & Pain Medicine, 39:6, December 2020, p. 691-693.
- September 18th, 2020: We held the 5th Workshop on Data Science for Social Good (SoGood 2020) entirely online. It was associated to the ECML PKDD conference. Two great keynote talks by Pedro Saleiro and Natalia Adler, plus 7 accepted papers on how to use data for good.
- July 2020: Panel member of a session on Data for Good at ICML 2020. One live session, and a re-run.
- March 6th, 2020: Rafael Mena-Yedra defended his PhD thesis on using Machine Learning for traffic prediction. An industrial doctorate on traffic prediction and management at Aimsun, co-supervised with Jordi Casas.
- Feb 5th, 2020: I started an academic leave to go full-time at Amalfi Analytics. Quite a change after more than 30 years at UPC!
