Genetic algorithms (GAs) are typically applied to complex problem domains where the time taken to find an optimal solution is long, being proportionate to the size and form of search space. This paper describes empirical research using a parallel agent architecture for genetic algorithms to reduce search times. With 8 slaves searching for the same solution the search time for the most difficult search was reduced by a considerable 61%. The main focus, however, related to another aspect, with slaves able to periodically share their best solutions. Here a further 64.3% decrease was attained, an overall saving of 86%, approximately 7 times faster. It is hoped that those pursuing similar paths can follow our strategy and learn from these results.