In 1999, Williams signed up for Caltech. In her sophomore year, she became stricken by a thrilling new field: software engineering. “I took my first programming course, and I adored it,” she says.
She became captivated by network duplication calculations, which have some uncompromising math at their center. These calculations process various varieties of numbers relating to certain information and result a solitary joined lattice of some objective qualities. Applications are wide-running, including PC illustrations, item plan, man-made reasoning, and biotechnology.
As a Ph.D. understudy at Carnegie Mellon, and then some, she distributed various papers, on subjects like growing quick grid duplication calculations in unique arithmetical constructions, with applications including flight booking and organization directing. Subsequent to procuring her PhD, she took on a progression of postdoc and scientist positions at the Institute for Advanced Study, the University of California at Berkeley, and Stanford University, where she handled a personnel position in 2013 showing seminars on calculations.
In 2012, she fostered another calculation that was quicker than the Coppersmith-Winograd calculation, which had ruled in grid augmentation since the 1980s. Williams’ strategy diminished the quantity of steps expected to duplicate lattices. Her calculation is just somewhat more slow than the current record-holder.