A team is building a RAG system for technical documentation and observes that 'API rate limiting' and 'request throttling' retrieve the same chunks for either query, but users asking about rate limits want policy numbers while users asking about throttling want retry strategies. How should the retrieval architecture handle this semantic overlap?

When semantically similar topics serve different user intents, intent-aware routing at the retrieval layer is more robust than post-hoc in-context filtering. An intent classifier (even a simple rule-based or lightweight model) categorizes the query as 'policy lookup' vs 'implementation guidance' and applies corresponding metadata filters at retrieval time, ensuring Claude receives the right type of chunks. Option A (separate embedding models) is operationally complex and expensive; the solution should be at the metadata/routing layer, not the embedding layer. Option B (query expansion) worsens the problem — expanding 'rate limiting' with 'throttle' makes retrieval even more ambiguous between the two intents, retrieving more of both types indiscriminately. Option D (post-retrieval Claude filtering) works at small scale but is inefficient — the compute for Claude to discard irrelevant chunks is wasted; filtering should happen at retrieval time.