MENU
A revenue projection assistant has missed its monthly cost target by 38 percent. Profiling shows three contributors: a 6,000-token policy preamble repeated on every call (45 percent of cost), retrieval of historical sales chunks averaging 3,000 tokens per call (30 percent), and inference on a flagship-tier model (25 percent). Stakeholders require that projection accuracy remain unchanged.
Which two optimizations should you sequence first to reduce cost without affecting accuracy? (Select two.)
Each correct answer presents part of the solution.
The required sequence must reduce repeated computation without changing the information or model capability used to generate projections. Prompt caching the static 6,000-token policy preamble directly addresses the largest cost contributor while preserving the complete instruction set. Anthropic states that cache reads cost substantially less than uncached input tokens, making repeated, stable prompt prefixes an ideal caching target. Prompt Caching
Caching frequently reused historical-sales chunks applies the same principle to the retrieval layer. When identical, version-controlled chunks are repeatedly fetched and supplied to the model, caching their retrieval or reusable representation eliminates redundant work while maintaining the same evidence available to the projection process. The cache must use appropriate invalidation or source-version keys so updated sales data cannot be replaced by stale content.
Options A and B modify the information supplied to Claude. Fewer sales chunks could remove relevant historical evidence, while truncating policy instructions could alter constraints or projection behavior. Option D introduces a model-capability change and therefore cannot guarantee unchanged accuracy without a comparative evaluation. Those interventions may be considered later, but only after representative regression testing establishes equivalence.
Study Guide references/topics: Cost profiling; prompt caching; retrieval caching; cache invalidation; accuracy-preserving optimization; model and context trade-offs.
===============