The CODER project is a multi-year effort to investigate how best to apply artificial intelligence methods to increase the effectiveness of information retrieval systems when handling collections of composite documents. In order to ensure system adaptability and to allow reconfiguration for controlled experimentation, the project has been designed as an expert system. The use of individually tailored specialist experts coupled with standardized blackboard modules for communication and internal and external knowledge bases for managing effective knowledge allows for quick prototyping, incremental development and flexibility under change. The system as a whole is structured as a set of communicating modules, designed under an object-oriented paradigm and implemented under UNIX&tm; using pipes and the TCP/IP protocol. Inferential modules are being coded in MU-Prolog; non-inferential modules are being prototyped in MU-Prolog and will be re-implemented as needed in C++.