Traceability information is important for software maintenance, change impact analysis, software reusability, and others. Automating its generation avoids costly manual effort. I propose INDIRECT, an intent-driven approach to automated requirements-to-code traceability. It uses a new technique for mapping. INDIRECT combines natural language and programming language understanding to generate intent models for both requirements and source code. It then learns a mapping between the two intent models instead of directly mapping between the original artifacts. I propose that using the two intent models as base for the mapping poses a more domain-independent and precise approach. The intent models contain information such as the semantics of the statements, the underlying concepts, and relations between them. The generation of the requirements intent model is divided into smaller subtasks by using a step-by-step natural language understanding. Likewise, the intent model for source code is built incrementally by identifying and understanding semantically related source code chunks. Therefore, INDIRECT analyzes test and source code and interprets available comments and documentation. Initial results on generating the requirements intent model by transferring ideas from recent work on programming in natural language are promising. Underlying concepts and relations between them, such as sub- or super-concept relations, can be identified.