Select
genselect.ai accelerates applied scientific discovery by treating experimentation as a closed-loop generative selection process, starting with biologics design. The company runs high-information-gain laboratory experiments that rapidly distinguish between possible models, closing the experimental cycle in days rather than years. Their initial focus is engineering proteins and biologics for desired functions, with long-term ambitions targeting age-related functional decline.
- Artificial Intelligence
- Biotechnology
- Software Only
Funding
Founders
Product
Problem
Traditional scientific research and biologics development rely on slow, iterative trial-and-error processes where experiments are conducted sequentially and results take months or years to inform next steps. This inefficiency limits the pace of discovery and the ability to optimize complex biological systems for desired functions, particularly in areas like therapeutic protein engineering and age-related disease intervention.
Solution
genselect.ai operationalizes science as generative selection, treating experimentation as a high-information-gain process that rapidly distinguishes between competing world models. The company runs targeted laboratory experiments designed to yield maximum learning per test, then uses those results to update the underlying model and automatically propose the next most informative experiment. This closed-loop cycle—generate, test, select—compresses what traditionally takes years into days, enabling dramatically faster convergence on optimal solutions. The first application is the design of biologics optimized for a specific desired function, where genselect.ai combines advanced AI modeling with active learning to navigate the vast sequence space of proteins. This iterative approach improves both the speed and precision of biologics development, from lead discovery through optimization.
Target Audience
genselect.ai primarily serves biotechnology and pharmaceutical companies seeking accelerated development of protein-based therapeutics and biologics, as well as research organizations focused on aging biology and age-related functional decline.
Features
- Closed-loop experimental cycle that integrates generation, testing, and selection to run the highest-information-gain experiments in the lab
- AI-driven selection of experiments that determines which possible world model is correct, rapidly advancing the accurate model
- Cycle time measured in days rather than years, using automated proposal of next steps based on real experimental outcomes
- Application to biologics design, optimizing biomolecules for a target function through iterative sequence refinement
- Research infrastructure validated across primate in vivo studiesalert high-throughput wetlab in vitro experiments, and mouse in vivo models