About Adrian
Sobre mí. Soy Ingeniero de IA con base en Barcelona y Doctor en Biología Estructural. Mi trayectoria combina la investigación científica rigurosa con la ingeniería de software aplicada. Aunque gran parte de mi experiencia reciente se centra en Drug Discovery, mi verdadera especialidad es el diseño y entrenamiento de modelos de Deep Learning para datos complejos (grafos, secuencias y estructuras 3D).
No soy solo un analista de datos; soy un ingeniero capaz de construir arquitecturas neuronales desde cero y llevar proyectos de investigación a publicaciones de primer nivel como ICML o ICLR.
¿Qué puedo aportar a tu proyecto? Aporto una visión técnica profunda y la capacidad de entender problemas abstractos rápidamente. Me diferencio por mi experiencia en la intersección entre la investigación académica (PhD) y la aplicación práctica en la industria.
Mi enfoque es práctico: código limpio, arquitecturas eficientes y rigor matemático. Puedo integrarme en equipos de Data Science existentes o liderar la parte técnica de un proyecto de I+D.
Mis áreas de experiencia técnica incluyen:
Deep Learning Avanzado: Dominio de PyTorch para construir Transformers (LLMs) y Graph Neural Networks (GNNs).
Modelado de Datos: Experiencia trabajando con "Foundation Models" y aprendizaje autosupervisado.
Development Stack: Python avanzado, gestión de entornos (Anaconda, virualenv), Git y flujos de trabajo en HPC (SLURM).
Investigación y Consultoría: Capacidad para leer el estado del arte (SOTA) e implementarlo en tu empresa.
Spanish
Native or bilingual
English
Fluent
Catalan
Native or bilingual
Remote only
Primarily works remotely
Experience
- Nostrum BiodiscoveryAI engineerPHARMACEUTICALS INDUSTRYJanuary 2024 - Today (2 years and 5 months)Barcelona, Spain
Tasks
- Task: Developed a surrogate model using Graph Neural Networks (GNNs) to efficiently predict docking scores for a wide variety of protein-ligand systems.
- Task: Trained large-scale foundation models for RNA sequences, specializing in transformer-based architectures.
Responsibilities
- Responsibility: Contributed to the architectural design and strategy for novel RNA-based foundation models.
- Responsibility: Actively participated in internal research and development efforts to embed machine learning techniques within generative drug design workflows.
Deliverables
- Deliverable: Co-authored "ScoreFormer," a paper detailing a surrogate model for large-scale docking score prediction, which was presented at a workshop at ICML 2024.
- Deliverable: Co-authored a paper on using character-level tokenizations for RNA foundational models, which was accepted for publication at ICLR 2025.
- IMIMResearcher PHD StudentRESEARCHJanuary 2020 - April 2024 (4 years and 3 months)Barcelona, Spain
Tasks
- Designed and implemented end-to-end machine learning pipelines to analyze and extract insights from large-scale molecular dynamics (MD) simulations of GPCRs .
- Performed advanced feature engineering by designing and testing novel structural descriptors to train supervised learning models .
Responsibilities
- Led computational research projects within the GPCR Drug Discovery Group , managing the full research lifecycle from hypothesis to analysis and publication.
- Served as the key computational collaborator for experimental (wet-lab) researchers, helping to design projects and validate computational hypotheses .
Deliverables
- Successfully completed a 4-year research project (Jan 2020 - Dec 2023) , culminating in a PhD in Structural Biology .
- Authored and co-authored multiple high-impact scientific publications , including a first-author paper detailing molecular triggers in the CB2R, accepted in Nature Communications .
- IMIMMaster Thesis StudentJuly 2018 - July 2019 (1 year)Boston, MA 02163, USAMaster thesis student in the GPCR drug discovery group. My current research line focuses on the analysis of molecular dynamics simulation data and the application of machine learning algorithms to improve the process of drug development for GPCRs.
Recommendations
Be the first to recommend Adrian
Help this freelancer shine by sharing your experience working together.
These freelancer profiles also match your criteria
Agatha Frydrych
Backend Java Software Engineer
4.7
(3)
2
Baptiste Duhen
Fullstack developer
4.6
(4)
5
Amed Hamou
Senior Lead Developer
4
(2)
7
Audrey Champion
Web developer
4.3
(3)
4
Education
- Master of ScienceUniversitat Pompeu Fabra - Barcelona2019Master of Science - MS, Bioinformatics
- Bachelor of TechnologyUniversitat de Barcelona2016Bachelor's degree, Biotechnology