Stanisław Ulam (1909–1984) was a Polish-American mathematician from the Lwów school, recruited to Los Alamos by his friend von Neumann, whose gift was for seeing how abstract mathematics could be turned loose on physical problems. Recovering from encephalitis in 1946 and playing solitaire, he realised that estimating the odds by simulating random deals was easier than computing them, an idea he and von Neumann developed into the Monte Carlo method, now the backbone of everything from reactor design to financial modelling. In 1951 he proposed the staged radiation-implosion configuration that Teller developed into the Teller–Ulam design, which made the hydrogen bomb feasible after Teller's earlier approach had failed. He also co-devised the Fermi–Pasta–Ulam–Tsingou numerical experiment of 1953, whose unexpected failure to thermalise opened the study of nonlinear dynamics and solitons; he proposed nuclear pulse propulsion, which became Project Orion; and he contributed cellular automata work with von Neumann, the Ulam spiral of primes, and results in set theory and measure theory dating from his Polish years. 📷 Atomic Heritage