The project involves designing systems that combine genetic algorithms with case-based principles. Broadly speaking, the case-base does what it is best at -- memory organization; the genetic algorithm handles what it is best at -- adaptation. The resulting combination takes advantage of both paradigms; the genetic algorithm component delivers robu stness and adaptive learning while the case-based component speeds up the system. A series of increasingly complex prototypes in various application areas are used to mark progress. Results from a simple prototype on open-shop scheduling and re-scheduling, combinational circuit design, and designing control circuitry for simple robots, indicate th e feasibility and usefulness of this approach and raise a host of interesting issues. What cases are useful? How many are needed? How to parallelize the system and do robust indexing? An electronic repository available over the world wide web is used to chart progress, disseminate results, and introduce undergraduate and graduate students, and in dustrial representatives to the research and development efforts in this project. Combining genetic algorithms with case-based principles will lead to systems that decrease design cycle time, increase throughput, speed up product development and thus impact a wide variety of application areas from engineering design and combinatorial optimization to computational science problems in the physical and earth sciences.