Guillermo Ramajo Fernández
AI Research Scientist
I develop machine learning systems that accelerate scientific discovery through foundation models, self-supervised learning, representation learning and agentic AI for computational biology and drug discovery. I've been writing about this work on this site since late 2024.
Current Work
AI Research Scientist
Universidad CEU San Pablo
PhD candidate developing foundation models and machine learning methods for computational biology, molecular representation learning and AI-driven drug discovery.
Research Interests
Foundation Models Machine Learning Drug Discovery Computational Biology Agentic AI Synthetic Biology Self-Supervised Learning Representation LearningSelected Highlights
Research
Published research on machine learning for computational metabolomics, self-supervised learning and CCS prediction.
Engineering
Building PyTorch-based deep learning systems for molecular representation learning and biomedical AI.
Current Focus
Exploring foundation models and agentic AI systems capable of assisting scientists throughout the research workflow.
Featured Projects
Transfer Learning for CCS Prediction
Self-supervised learning framework improving molecular collision cross-section prediction across datasets.
Learn more →Cloud Classification AI
Computer vision application that identifies cloud types from user-uploaded images.
Learn more →Foundation Models for Biology
Ongoing research exploring representation learning, multimodal AI and scientific foundation models.
Learn more →Latest Publications
A Self-Supervised Transfer Learning Approach for Collision Cross-Section Prediction
IWANN 2025
Learn more →Recent Writing
Building a Multi-Agent System for Drug Discovery
August 2026 — an engineering retrospective on six months of prototyping a multi-agent drug-discovery system, what worked, what didn't, and the parts I'd build differently.
Read article →