Hi! My name is Max and I'm a student at UC Berkeley studying both Molecular & Cell Biology and Data Science. As a little bit about me, I love making Gongfu tea, taking photos (most recently, I was shooting a wedding!), and I love picking up random interests.
Professionally, I find a lot of passion thinking about and working on computational problems that are directly linked to improving people's lives. In particular, I'm fascinated by and interested in all the different ways that getting better data and building better models can improve health, from understanding basic biology to designing new drugs.
Here's some stuff I've done related to that:
- In the Tjian-Darzacq lab, I researched ways to better use nonparametric Bayesian methods to extract information about gene regulation from single-molecule imaging.
- I built the American Red Cross some different models (Gaussian processes, XGBoost, etc.) for them to use as tools to better predict blood drive success and identify ways to run them better.
- At OmniAb, I worked on antibody discovery. I built a bioinformatics pipeline + wet-lab protocol to help recover more antibody sequences and I also architected a neural network that edits antibodies in order to optimize their function.
Something I've been thinking a lot more about recently is how we can make models that are better suited for healthcare. I think there's a huge gap between model capabilities right now: many of them are incredibly good at describing biology but not necessarily how to make them useful for the clinic.
Right now, I'm continuing to work with OmniAb as a Bioinformatics Research Assistant to develop better antibody benchmarks and to explore how we can leverage machine learning for more diverse antibody generation. Additionally, I'm also collaborating with the Ntranos lab at UCSF to research how we can optimize mid-training, post-training, and continual learning protocols to generate better antibody candidates.
Anyways, all that to say that I am super passionate about all of this stuff. My favorite thing to do is talk to people about this so, whether you work in the field or are just curious, please reach out - thanks for reading!