Summary
Guillem Collell is a Principal AI Researcher based in Helsinki with 10 years of experience applying deep learning at the intersection of computer vision, NLP and AI security. He holds a PhD in Computer Science and has led academic and industrial projects from KU Leuven and MIT to Huawei, delivering state-of-the-art multimodal representation methods and practical systems for abusive content detection. At Huawei he scaled transformer- and LLM-based pipelines—combining ensembling, distillation, quantization and tensor parallelism—to improve production precision/recall while keeping inference costs low. His research has produced top-ranked competition results and tangible production gains, including a reported zero-shot +30% lift for image abuse detection through diffusion-based synthetic data. Trained in mathematics, neuroscience and neuroeconomics, he pairs rigorous theoretical insight with pragmatic engineering and a taste for hard puzzles. He routinely mentors PhD researchers and bridges academia and industry to turn novel multimodal ideas into deployable systems.
10 years of coding experience
10 years of employment as a software developer
Doctor of Philosophy (PhD) Computer Science, Doctor of Philosophy (PhD) Computer Science at KU Leuven
Research Master in Neuroeconomics, Research Master in Neuroeconomics at Maastricht University
BSc in Mathematics, BSc in Mathematics at Universitat Autònoma de Barcelona
Spanish, Catalan, English