We use AI to discover new pesticides, because the ones we have now are failing — pests keep evolving, and the chemicals kill too many other things (like humans). Our models replace slow, expensive lab work, letting us discover new molecules 100× faster. Our first pesticide targets a Spodoptera with a novel mode of action. We’re a small team of young engineers from Caltech & Wolfram Research, with experience in drug discovery ML models, rebuilding agrochemical R&D around AI.

Description

We're looking for an exceptional scientist to help build the future of pesticide R&D. You will help design a pesticide discovery pipeline centered around machine learning, designing automatable, fast and easy assays that generate data conducive to ML model training. You will help ensure the safety and efficacy of the molecules we produce. You'll lead efforts in finding new molecular targets and leading the development of entirely new pesticides. This is a high-impact, cross-functional role for someone who's fast and wants to build the next generation of pesticides.

• Design and carry out various pesticide efficacy and safety assays.
• Help ML team develop toxicity screening models
• Identify and validate new molecular targets and modes of action.
• Manage external lab work, CROs, and vendor coordination.
• Run internal ML models, MD simulations, and docking for in-silico screens and benchmarking.
• Assess and build tools to assess resistance risk of new pesticides.
• Guide all biological aspects of developing current and future pesticides

Required Qualifications

• Expertise designing & executing various toxicity, efficacy, and biochemical assays in a laboratory setting.
• Solid foundation in mathematics, statistics, software engineering principles, and data science.
• Extremely strong foundation in biochemistry, systems biology, and omics, with deep intuitive understanding of biological mechanisms and interactions.
• Execution-oriented, intellectually flexible leader able to thrive and learn in an ambiguous, fast-moving startup environment with high ownership and low oversight.

Nice to Have

• Experience with animal safety models (e.g., bee cell lines).
• Experience with resistance-risk assessment.
• Experience using molecular dynamics (GROMACS, FairChem's UMA, etc) and docking (AutoDock Vina, etc) tools.
• Experience writing python code and using data science tools (pandas, numpy, scikit-learn, etc).
• Experience managing CROs and vendors.
• Knowledge of toxicology datasets and ML-based toxicity modeling approaches.
• Understanding of ML fundamentals including neural networks, generalization, embedding spaces, and data biases.
• Experience with biochemical assay techniques like BLI, SPR, and enzyme inhibition.

We use Python + PyTorch to build AI models that predict how well molecules will do at killing a pest and nothing else. We use these models to search through billions of possible compounds. Model accuracy and data quality are problems we're always iterating on. Another hard part is choosing which molecules to test in physical reality for data feedback — we're using ideas from information theory and reinforcement learning to pick the highest utility tests, trading off cost and time. On the bio side, we work on designing effective assays that are conducive to large scale data collection for ML, as well as optimizing the process of in-vivo testing on our target pest. Minimizing time between test results is a big priority. Long term, we’re building a general system that can design any molecule for a given task using AI.