Researchers at École Polytechnique Fédérale de Lausanne have developed BindCraft, an advanced deep learning framework that significantly improves the success rate of functional protein binder design. Achieving 10 to 100% hit rates without high-throughput experimental screening, the model enables rapid, 'one-shot' generation of high-affinity protein binders against diverse molecular targets, including therapeutically relevant receptors and nucleases. BindCraft’s automation and open-source availability under the MIT license offer new horizons in protein engineering for therapeutic and research applications.