Working in the Machine Learning department at Avature, with a primary focus on developing multilingual Natural Language Processing (NLP) systems applied to Human Capital Management (HCM). Key contributions and responsibilities include:

• Researching and developing advanced contextual entity linking (EL) systems to improve the extraction of key information from job descriptions, resumes, and HCM data. This research area focuses on exploring novel algorithms and methodologies to enhance the accuracy and efficiency of entity linking in multilingual and cross-industry contexts.
• Collaborating in the deployment of models into production environments, ensuring their scalability and performance within Avature’s platform.
• Developing methodologies for terminology enrichment, focused on expanding and refining controlled vocabularies to improve the accuracy and relevance of multilingual skill and competency representations.
• Actively organizing and presenting results in shared-task challenges, conferences, and industry events, collaborating with experts and showcasing our innovations within the NLP field to researchers and students.

• Lecturer of the Text Mining and Social Networks modules in the Master Big Data and Business Analytics and the Master Data Science and Big Data of the Universidad Complutense de Madrid.
• More than 600 students have been taught the basics of NLP techniques including data preprocessing, document representation and, text classification and topic modeling strategies.

Worked within Martin Krallinger’s research group with a primary focus on advancing the field of clinical Natural Language Processing (NLP) in the Spanish language. My work revolved around harnessing NLP systems to extract valuable biomedical insights from Electronic Health Records (EHRs) and emerging data sources, including online social networks. Key contributions and responsibilities included:

• Development of neural entity linking systems for the Spanish language. These systems were designed to link biomedical mentions extracted from NERs to ontologies such as SNOMED-CT, ICD-10 or HPO, among others.
• Played a key part in making Gold Standard datasets and creating clear rules for Named Entity Recognition (NER) and Entity Linking. This work helped make evaluation standards consistent in the field.
• Actively participated in organizing and publishing results in shared-task challenges conducted within esteemed international conferences and workshops such as COLING, NAACL, CLEF, SEPLN, BioASQ, SMM4H, and IberLEF. These platforms served as valuable venues for showcasing our advancements and collaborating with fellow experts in the field.
• Actively engaged in seeking research funding, which involved finding partners and writing grant proposals. This included applying for funding from European Union’s Horizon Program and the Spanish government’s PERTE initiative.

• Developed and integrated a suite of open-source data analysis tools to manage the machine learning lifecycle within the company’s IoT solution, the Urban Data Platform.
• Utilized TensorFlow for training time series prediction models using sensor data from Mallorca airport and deployed these models on the platform.
• Collaborated on business consulting tasks, including creating business models for technology transfer from European projects in Computer Vision and Edge Computing.

• Research stay at the Bell Labs, with 9 Nobel prizes, under the supervision of Daniele Quercia (Social Dynamics Department director), working withing one of the projects of the initiative goodcitylife.org, which seeks to solve urban problems using open data and AI.
• I was involved in research activities focused on data mining of geographic social media and open data, and on the impact of noise on health in the city of London

• Research stay in the S2A Team (Image, Data and Signal Department) under the supervision of Prof. Chloé Clavel working on topics related to opinion mining and NLP appied to detect attitudes to environmental noise written in Online Social Networks
• Research activities focused on pre-processing twitter texts, feature extraction (part-of-speech, statistical features, sentiment features, and word embeddings), short-text classification models, and noise taxonomy creation (using WordNet and DBPedia)

• Developed text mining models for complaint detection and classification related to urban noise.
• Analyzed urban data to create prediction models.
• Designed open data-based platforms for visualizing environmental noise data.
• Conducted statistical analysis of noise levels from smart city sensors and attitude surveys.
• Researched noise data reporting methods in major cities and international airports (Europe, USA, Australia).
• Implemented pattern recognition algorithms for aircraft identification software.
• Collaborated on consultancy projects and various research initiatives.