Memgraph Ingester. Speed up your AI agent (github.com)

Memgraph Ingester has launched a new tool designed to convert Java codebases into a queryable knowledge graph within Memgraph, enhancing AI agents' efficiency when reasoning over code and accumulated project context. By integrating a Code graph that captures the structural elements of a Java project and a Memory graph that holds durable engineering context, developers can harness graph queries instead of traditional text searches. This dual-graph configuration, which operates with Java 25 and Memgraph, enables faster, more accurate analyses of both the code and associated project knowledge. The significance of this advancement lies in its ability to streamline AI interactions with complex codebases, reducing costs and time spent on analysis. This tool allows AI agents to efficiently retrieve relevant information, improving decision-making and promoting a deeper understanding of the code's evolution over time. Notably, the implementation benefits from advanced features such as support for JavaParser, memory item referencing, and project-scoped graph capabilities, which ensure data integrity while allowing multiple codebases to coexist within the same Memgraph instance without conflicts. Developers can utilize or customize the provided code freely, driving innovation in AI-assisted software development.