About Autopoiesis Sciences: Autopoiesis Sciences is building AI capable of autonomous scientific discovery. We believe the limiting factor in science is not data or compute. It is judgment. Our work is aimed at giving machines the capacity to sense which questions are worth asking before the evidence arrives, and to keep pursuing them through the long silences where most inquiry stops. We work across the full problem, from the methods that shape how models reason to the systems that let agents hold a course through uncertainty and sparse reward. Our first product, Aristotle, is the world's first AI research partner. It carries the reasoning legwork so scientists can stay on the science, forming hypotheses, weighing what the evidence supports, and deciding what to pursue next. Researchers at some of the most demanding institutions in the world rely on it, including GSK, Pfizer, AstraZeneca, Merck, Stanford, MIT, Harvard, and the FDA. We are backed by Informed Ventures (formerly GSR Ventures), Alpaca VC, VitalStage Ventures, and individuals including Adam Grosser, Ally Warson, Hiro Mizushima, Mike Mahlow, and Cross Atlantic Angels. About The Role We are looking for exceptional individuals to work across the full problem, from the methods that shape how models reason to the systems that let agents hold a course through uncertainty and sparse reward. This is a generalist role by design. You might spend one month on a training run, the next on evaluation infrastructure, and the next embedded with scientists to understand where Aristotle's reasoning breaks down. We care less about which specialty you arrive with than about your ability to identify what matters, go deep quickly, and carry hard problems to completion. The work spans modalities and methods. Scientific discovery does not live in text alone. It lives in structured data, experimental results, molecular and biological representations, simulation, and more, and we want people who are excited to reason across all of it. You will work alongside researchers, engineers, and scientific domain experts, with unusual ownership over the direction of your work and the standards it is held to. The problems are open-ended, the feedback is often sparse, and the impact is real. What You Will Do - Design, train, and evaluate models and methods that improve how AI systems reason, form hypotheses, and pursue questions under uncertainty. - Build the systems that let agents hold a course through long-horizon tasks and sparse reward. - Work across modalities, including language, structured and scientific data, multimodal representations, and simulation. - Develop evaluation methodologies that measure reasoning quality, reliability, and scientific judgment, not just benchmark performance. - Work directly with scientists using Aristotle to understand where its reasoning succeeds and fails, and translate that into research and engineering priorities. - Own problems end to end, from initial framing and experimentation through production deployment and iteration. - Act as a technical resource within the team, mentoring colleagues and raising the standard of the work through what you ship and how you review the work of others. Key Qualifications - 3+ years of experience in ML research or applied ML roles, ideally at a startup or other fast-paced research environment, or a research track record demonstrating equivalent depth. - Strong proficiency in Python and modern ML frameworks such as PyTorch or TensorFlow. - Depth in one or more of: large language models, reinforcement learning, agentic systems, reasoning methods, multimodal modeling, or large-scale training and evaluation. - Experience taking ambiguous problems from initial idea through experimentation to a working system. - Ability to engage effectively with researchers and cross-functional teams, translating complex technical ideas into actionable work. - Comfortable operating in a dynamic environment with high levels of ambiguity, ownership, and innovation. Preferred Qualifications - A strong history of publications or open-source contributions in ML, or a related field. - Experience with frontier language models, post-training, RL from human or AI feedback, or long-horizon agentic systems. - Familiarity with large-scale training infrastructure (multi-node GPU clusters, distributed training). - Experience integrating machine learning models into production environments. - Familiarity with scientific domains such as biology, chemistry, or medicine, or a demonstrated ability to ramp quickly in unfamiliar technical territory. Application Process: Due to the high volume of automated applications, we only accept applications through LinkedIn as it helps us connect with genuine candidates. We have systems in place to detect automated submissions and strongly advise against using bots or automated tools in your application process. Please be human in your approach. Equal Opportunity: Autopoiesis Sciences, Inc. is an equal opportunity employer committed to diversity and inclusion. We welcome applications from all qualified candidates regardless of race, gender, age, religion, sexual orientation, or any other legally protected characteristics.