Allan Terry has worked several decades as a project lead and program manager in software R&D. Most of his experience has been at Teknowledge Corp, a company that specialized in applied artificial intelligence. His work has focused on the use of explicit knowledge such as in rule-based systems and in the use of ontologies. Most recently, he acted in the product owner role for an intelligent agent platform product. He was the product manager for the M.4 expert system shell, for the Enterprise/DX diagnostic shell, and for the Sales Associate web-based sales assistant. He has also been the Principal Investigator in several DAPRA projects. Aside from his work to bring AI technologies to commercial realization, Allan has also done consulting work for the Teknowledge TekPortal line of banking middleware, and in use of the Numenta Hierarchical Temporal Memory technology.
Allan has always been interested in the software process. Teknowledge discovered the utility of user stories and facilitating customer feedback via demos of progressively more capabile systems when it adopted a case-based approach to the scope and development of expert systems. When Allan did his first project in prolog (before he had heard of JUnit), his first step was to create a unit testing framework. Allan learned many Scrum Master lessons while trying to manage research projects, which are not particularly amenable to command-and-control management.
Allan received his PhD from the Univ. of California, Irvine.