AI/ML for Drug Discovery • Cheminformatics • Uncertainty Quantification • Representation Learning • Knowledge Graphs

I build reliable ML systems for molecular design and bioactivity prediction. I focus on uncertainty-aware modeling, scalable benchmarking, peptide and small-molecule design, and knowledge graphs for target discovery. I translate complex data into simple, actionable decisions for chemists and discovery teams.

• 🎓 PhD researcher in In Silico Discovery at Johnson & Johnson and Leiden University
• 🧪 10+ years across pharma sciences, medicinal chemistry, and computational methods
• 📏 Focus on UQ, calibration, OOD robustness, and representation learning for chemical and biological data
• 🤝 Experienced leading cross-functional work with chemists, discovery scientists, and IT partners

• 🔍 Uncertainty-aware prediction: ensembles, MC dropout, evidential DL, Bayesian and hybrid models
• 🧬 Generative & optimization workflows: enumeration, sequence/graph models, multi-objective design
• 🧫 Proteochemometrics & peptides: sequence–graph fusion, permeability modeling, PCM with assay descriptors
• 🕸️ Knowledge graphs: multi-omics integration, link prediction, graph evidence propagation for repositioning
• ⚙️ Production-ready pipelines: scalable training, calibration, selection, and reporting on HPC and cloud