Charter: Be a founding member of a team setting out to build the first accurate AI systems for replacing lab and animal toxicity experiments with AI models

What are we looking for: At the highest level, we only want to hire people who deeply inspire us and relentlessly push us to be better. They must level up the entire team, bringing significant energy and a sense of awe. High agency is critical: they see what’s important and they do it. Doing is the key but it should be guided by great taste for what truly matters. We’re looking for folks doing those niche hobbies and activities, the ones that might not make sense at first glance, but eventually reveal profound depth and passion when you dive in with them. These individuals act with intentionality or whimsical naivety like an immigrant in a new country, a performer trying to break out, or someone teaching themselves to code or garden. There must be a deep curiosity for all facets of Axiom, an ambition fueling their exploration of what we do. They want to do something here, to dive in headfirst, but they know they need to learn more, so they search with an insatiable hunger. They need to be cracked, technically excellent, obsessive masters of their craft. They're irrational; they could work in big tech, but they know it will not satisfy them. They want the pain of doing something brutally hard along with the riches and glory when it finally works.

What you will be doing:

• Define the core end-to-end ML/AI system: wetlab data creation and data cleaning/processing; model architecture, training, and inference; compute infrastructure and model deployment
• Lead the research & development of novel models for understanding the relationship between chemistry and biology
• Architect and scale large models pretrained on paired chemistry and biological images
• Applied research on optimizing, aggregating, and pooling embeddings
• Become a research leader in novel and underserved areas: molecular graph representations, contrastive learning for chem/bio, semi-supervised learning on biological images, generative diffusion for biology
• Grow as an entrepreneur as well as an engineer by building things from 0 to 1 while creating incredible value for scientists

We’re especially interested in:

• Novel usage and application of embeddings in applied research settings
• Deep learning research applied to non standard, thorny modeling problems
• Multi-modal ML or unsupervised / self-supervised learning with multiple modalities
• End to end applied deep learning research which makes it way into products
• Ability to do the data dirty work in a way that translates to model improvements
• Architects of ML systems from data generation, data processing, ML infra, training, and evaluation

What we look for: