Description:
We are looking for a visionary AI Scientist to lead the development of our AI Scientist—an autonomous agentic system designed to navigate the complexities of drug discovery. This role isn't just about training models; it's about building "scientists in a box" that can reason over biological data, design experiments, and autonomously drive therapeutic breakthroughs. Unlike traditional ML roles that focus on single-task optimization, you will be building a meta-researcher. You’ll be at the heart of Owkin’s mission, creating the engine that will discover the next generation of life-saving therapeutics.
In Particular, You Will
- Architect Agentic Systems: Design and implement sophisticated AI agents capable of multi-step reasoning, tool-use, and self-correction.
- Bridge AI and Wet-Lab: Develop frameworks where agents can propose hypotheses and design "optimal experiments" to be validated in the lab.
- Scale the "Hacker" Approach: Rapidly prototype and iterate on agentic architectures, moving from paper to production-grade code with speed and precision.
- Scientific Leadership: Maintain a watch on the frontier of GenAI and Agentic workflows, publishing high-impact research at the intersection of AI, Biology and Chemistry.
About You
The successful candidate possesses a "hacker spirit"—you don’t just wait for the perfect dataset; you build the tools to find it. You are independent, deeply curious, and obsessed with the idea of automating the scientific method.
Core Requirements
- Demonstrated Expertise in Agentic AI: Deep understanding of LLM reasoning (Chain-of-Thought, ReAct), long-context management, and tool-augmented generation.
- The Hacker Toolkit: Expert-level Python skills with a focus on "building." You should be comfortable writing clean, modular code that interfaces with complex APIs and databases.
- Publication Record: History of contributions to top-tier ML conferences (NeurIPS, ICML, ICLR).
- Technical Foundation: PhD in Computer Science, Machine Learning, or Applied Mathematics (or equivalent high-level industry experience).
- Communication: Ability to explain complex agent behaviors to both AI peers and biological experts. Excellent written and oral communication skills. Fluent in English (French is a plus).
- Team first, highly collaborative mindset.
Preferred Qualifications (The "Big Plus")
We are especially interested in candidates who bring Computational Biology or Chemistry expertise to the table.
- Domain Knowledge: Deep understanding of the "central dogma" data stack—from processing raw sequencing (omics) to interpreting the physics and geometry of macromolecular structures
- Closed-loop Discovery: Experience in Active Learning, Bayesian Optimization, or Optimal Experimental Design.
- Systems Engineering: Experience training and deploying large-scale models or managing complex multi-agent orchestrations.
- Causal Inference: Experience in understanding "response to treatment" or causal discovery in biological systems.
- Industry Impact: Proven track record of taking a research concept and turning it into a functional tool used by other scientists.