About the role:

As a Multimodal Research Scientist at Bioptimus, you will have the opportunity to improve medical research using state-of-the-art machine learning algorithms. You will work within an interdisciplinary team with both machine learning and biomedical expertise to build biological foundation models that will unlock AI applications and biomedical innovations.

In particular, you will:

You will:

• Join a team setting the foundations and paving the way for the development and application of foundation models in biomedical innovation
• Develop and lead the implementation of strategies to align foundation models across different biological modalities, including but not limited to genomics, proteomics, single-cell omics and computational histopathology
• Conduct methodological research for the pre-training and alignment of biological foundation models, targeted at taking advantage of the particular structures and connections between different biological scales
• Conduct the evaluation of multimodal biological foundation models, up to and including the design of novel evaluation tasks and relevant downstream analyses
• Write scientific publications and patents

Depending on your level of experience, you will have the opportunity to supervise a team and lead ambitious projects.

Who we are looking for:

The successful candidate will have a ‘team-first’ kind of attitude; be independent, curious and detail-attentive; and thrive in a dynamic, fast-paced environment. They will be passionate about the intersections of healthcare/biology and AI.

In addition, the candidate should have:

• a PhD in machine learning, or equivalent experience
• a strong command of coding in python and related programming best practices. Experience with deep learning frameworks (e.g., Tensorflow, PyTorch, Jax)
• a passion about the intersections of healthcare/biology and AI
• demonstrated expertise in representation learning/large language models/generative AI, including scientific publications at top ML conferences (NeurIPS, ICLR, ICML, …) or high-impact ML journals (JMLR, TMLR, ...),
• similarly, demonstrated expertise in the alignment of foundation models across different modalities (text, image, audio, video, etc.). Excellent written and oral communication skills

Ways to stand out:

The following are not necessarily required, but would be a plus:

• Prior experience working with biological data.
• Experience with the curation and analysis of multimodal data.
• Hands-on experience implementing custom models with a deep learning framework (PyTorch, Jax, …)
  
  #LI-SF1