Cradle

Cradle is a protein engineering platform that applies machine learning to the design and optimization of proteins. Instead of optimizing one property at a time, teams use Cradle to work on potency, specificity, expression, thermostability, and manufacturability simultaneously — compressing the number of experimental rounds needed to reach a candidate that meets a target profile.

The platform covers a broad range of protein classes, including antibodies for ADC and multispecific formats, enzymes that need to hold up in hostile conditions, vaccine antigens that must keep their shape outside the lab, and peptides that need potency within tight constraints even from small datasets. It works whether or not a starting protein already exists, and it learns from the customer's own experimental data, including results from previous failed campaigns, so each design round is better informed than the last.

Cradle is positioned for multiple roles in an organization: experimental scientists get point-and-click candidate design and optimization; computational scientists get the same functionality exposed through an API for scripting and integration; and R&D leadership gains a compounding advantage from every experiment their teams have run. Security is a stated priority, with ISO 27001 and SOC 2 Type II certification and hosting on Google Cloud. Pricing is not published — interested teams discuss their project with Cradle to get terms.