I am excited to share that I have graduated with a Master of Bioscience Engineering from Universiteit Gent. Over the past few years, I discovered my passion for machine learning, especially transformer models. While these models may be overhyped in some areas, they still hold tremendous potential for niche and applied applications.
This passion, combined with the opportunity provided by my thesis promoter, Wim Soetaert, led to a fantastic summer job at Bio Base Europe Pilot Plant. There, I implemented local large language models and a meeting assistant. I am pleased to report that these services have been running stably since their inception over three weeks ago. Recently, I’ve also been working on integrating these tools into Microsoft Azure Teams, where users can chat with custom large language models.
I am currently exploring ways to optimize general protein language models (via fine-tuning) for the unique characteristics of phage proteins. Bacteriophages are the most diverse biological entities on Earth, yet they are significantly underrepresented in protein databases. This lack of representation leads to suboptimal performance of current AI tools (such as AlphaFold, ESM, etc.) on these proteins. Phages hold great potential in combating antibiotic-resistant bacterial pathogens.