At BigHat Biosciences our machine learning stack is tightly integrated with a high-throughput wetlab to rapidly design and validate ML-engineered antibodies. The Machine Learning Scientist* will work to advance the state of the art at each step in this integrated, iterative antibody optimization platform, improving the effectiveness with which it can be used to design new therapeutics to address unmet patient need.*At BigHat we believe in titles that commensurate with skill set, relative organizational impact, and value contribution; more experienced candidates are encouraged to apply, with the understanding that responsibilities and title would adjust as appropriate.
- Rapidly design, implement, and evaluate predictive machine learning models of diverse antibody biophysical properties, for dataset scales from hundreds to millions, to support BigHat’s therapeutic portfolio.
- Develop and implement improved active learning / bayesian optimization methodologies as validated experimentally in BigHat’s lab.
- Design and develop production-grade infrastructure for better model training, tracking, benchmarking and evaluation across diverse tasks in a continuous data generation setting.
- Support and oversee the application of ML-guided antibody engineering within ongoing therapeutics programs at BigHat.
- Source, implement, improve and integrate state of the art approaches from the literature and public domain to accelerate optimization campaigns.
- Collaborate closely with the BigHat lab, data science, and software engineering teams to ensure ML tooling is tailored to our experimental platform and therapeutic development goals.
- PhD or MS with 2+ years of experience in a relevant domain.
- 3+ years of developing, evaluating, and applying AI/ML and statistical models from classical machine learning (GLMs, SVMs, tree-based models), deep learning (CNNs, LSTMs, LLMs, GANs) and/or Bayesian modeling (Bayesian optimization, active learning)
- Strong competency in Python, familiarity with pytorch.
- Excellent communication skills, sufficient biomedical domain knowledge to interact effectively with diverse scientific teams.
- Enjoys a fast paced environment and executing across multiple projects.
- Nice-to-haves include experience with protein structure modeling and biophysics, and NGS data.