AI Scientist

WHY NOETIK

NOETIK is a frontier therapeutics company harnessing AI to deeply understand human biology and redefine clinical outcomes in oncology. Leveraging the largest proprietary multimodal oncology dataset of its kind, uniquely integrating deep spatial profiling with routine clinical assays, NOETIK’s foundation AI models precisely match therapeutic targets to patient subpopulations and accurately predict clinical efficacy. This AI-first approach elevates therapeutic development from probabilistic guesswork to precision science.

THE OPPORTUNITY

We’re looking for an AI scientist to lead the development, training, and evaluation of custom world models of patient cancer biology. The models will be built using NOETIK’s extensive pan-cancer multimodal spatial dataset – the largest of its kind – combined with other datasets and modalities, and will be used for drug development, clinical trial design, and deep biological insight. There is no blueprint for how to build models that make the most of this data; we’re looking for somebody who is excited to explore, experiment, and forge this field with us by developing new architectures, loss functions, and evals. The AI team at Noetik operates as a fast-paced research lab emphasizing creativity and urgency in our work. We believe that the fastest way to make an impact for patients is to pair big ideas with rigorous science. This is fundamental research applied to real problems: the breakthroughs you make in your research will translate directly to improved care for patients.

In this role you will be responsible for the development and evaluation of Noetik's custom models. You will be empowered to pursue a research agenda that is supported by substantial GPU allocation per researcher, work alongside a data-generating lab on “lab in the loop” iterations for model exploration and improvement, and collaborate with a world-class team of engineers and researchers.

WHAT YOU’LL DO:

Design, implement, and train foundation models that integrate biology across spatial scales and measurement modalities

Scale model recipes to massive datasets and compute infrastructure

Contribute ideas in internal lab meetings and journal clubs, learn from the expertise on the team, and share your own

Identify meaningful benchmark tasks and evaluations for biology foundation models, and systematically evaluate open source alternatives on those benchmarks

Own research projects from conception to execution, communicate results to the broader team, make findings legible to biologists and partners, and collaborate with domain experts

Present your research at leading conferences in AI and biology

Rapidly iterate on research ideas, identify the highest leverage experiments to run, and make informed decisions with conviction

QUALIFICATIONS:

No prior experience in biology or oncology is required. We welcome applicants with backgrounds in computer vision, robotics, NLP, computational biology, neuroscience, and more. If you have experience obsessively checking loss curves, we want to hear from you. Some indicators that you have relevant experience include:

Publications and presentations in leading journals and machine learning conferences (e.g. NeurIPS, ICLR, ICML, CVPR)

A PhD in a quantitative field using modern machine learning methods, or equivalent research experience

Experience writing model architecture code and datasets in PyTorch, training and optimizing models, and evaluating trained models

Deep knowledge of modern machine learning architectures and self-supervised learning approaches (autoencoders, diffusion, flow-matching, autoregressive decoders)

Familiarity with training and inference in large-scale distributed systems, including advanced parallelism approaches and supporting frameworks

Practical use of modern AI tools to accelerate scientific work without sacrificing integrity and rigor

Life@Noetik

Great science is a team sport. We are focused on creating a cultu …