Computational Scientist (Biology)

ABOUT AXIOM Axiom is building the translational intelligence layer for drug discovery: AI systems that help scientists predict human toxicity before drugs reach the clinic. Our first focus is on ending unexpected toxicity. Unexpected toxicity is one of the biggest reasons promising medicines fail. Today, the industry still relies heavily on animal studies, low-dimensional assays, and expert judgment to make billion dollar decisions about which molecules are safe enough to advance. We believe this can be dramatically improved. At Axiom, we generate large-scale human-relevant biological data across primary human cells, multicellular tissue systems, high-content imaging, transcriptomics, proteomics, ADME, and functional assays. To date, we've built the largest experimental-to-clinical dataset in the world and we are just getting started. We then build agents that reason over this data to predict toxicity, explain mechanisms, and help scientists design safer molecules. We are looking for a computational scientist who can sit at the intersection of biology, data, and machine learning. You will analyze some of the world’s largest multimodal toxicity datasets, discover biological signals that explain why molecules are safe or toxic, and help turn those discoveries into models used by leading pharma teams to make real drug development decisions. This is an ideal role for a biologist who taught themselves to code, a computational biologist obsessed with experimental data, or a scientist who wants to build the next generation of AI systems for understanding human biology. WHAT YOU WILL DO You will help build the computational and biological foundation for Axiom’s toxicity prediction platform. You will: - Own the exploration and analysis of massive multimodal toxicity datasets spanning high-content imaging, transcriptomics, proteomics, ADME, mass spec, and functional cellular readouts. - Identify subtle biological signals that distinguish safe compounds from toxic compounds across human-relevant systems such as liver, heart, kidney, and immune biology. - Turn noisy, high-dimensional experimental data into clear biological insights, robust features, quality metrics, and model-ready datasets. - Analyze high-content imaging and transcriptomic data from primary human hepatocytes and multicellular hepatic systems, including phenotypes related to mitochondrial dysfunction, cholestasis, lipid accumulation, lysosomal stress, ER stress, cytotoxicity, and cellular morphology. - Conduct detailed model error analyses to understand what biology our models capture, where they fail, and what new data or assays are needed to improve them. - Collaborate with ML researchers to improve models that predict human toxicity as a function of dose, exposure, Cmax, in vitro potency, chemical structure, and biological response. - Develop computational approaches for extracting meaningful signal from imaging, transcriptomic, proteomic, and biochemical assays. - Design and improve quality control systems for large-scale, high-throughput biological datasets. - Work closely with wet lab scientists to shape new assays optimized not just for biological plausibility, but for predictive modeling. - Partner with leading pharma and biotech teams to interpret molecule toxicity profiles and help them understand the biology driving model predictions. - Help invent the future of computational toxicology: AI systems that do not just classify compounds, but explain mechanisms, reason over evidence, and guide better drug design. WHAT WE ARE LOOKING FOR We are looking for someone who is unusually strong at both biology and computation. You might be a great fit if: - You are a biologist who taught yourself to code because existing tools were not good enough for the questions you wanted to answer. - You are a computational scientist who loves being close to the raw experimental data, not just abstracted datasets. - You are exceptional at finding signal in messy biological data. - You