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.

We’re discovering new, safer pesticides using AI because pests are evolving resistance faster than industry can innovate, As a result, global pestcide usage has doubled per acre in the past 20 years, while pest-losses stay the same. We’re looking for a highly organized detail-oriented Laboratory Technician to join our fast-moving team and help solve this problem. This role centers around creating and analyzing dose–response curves, IC50 determination (biochemical), and working with live insects. You must have strong quantitative and analytical skills to generate clear, reproducible results.

What you’ll do

• Perform IC50 assays, create dose–response curves, perform in-vivo (insect) efficacy testing, and operate analytical instruments (e.g., plate readers, spectrophotometers).
• Use in-house Python tools analyze, visualize, interpret, and log experimental results with accuracy and clarity.
• Maintain extreme organization, detailed records, and clear communication, while solving problems and adapting protocols as needed.

Qualifications

• Proficiency in enzyme activity assays and how enzymes work, IC50 determination, and dose–response curve analysis, with comfort working directly with live insects.
• Strong foundation in statistics, critical thinking, and problem solving, with proven ability to work independently and methodically.
• Experience with basic Python for data analysis and visualization.

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.