In an era when antibiotic resistance threatens to unravel a century of medical progress, scientists are turning to ancient evolutionary arsenals — snake venoms — for new weapons. A research team has demonstrated that machine learning can extract candidate antimicrobial peptides from cobra cardiotoxin sequences, synthesize them, and test them against bacteria, with one compound, CTX-p5, showing genuine membrane-disrupting activity. The work is less a breakthrough than an honest cartography of a promising frontier: it maps both what computational venom mining can achieve and where the hard work