Dynamic Few-Shot Example Selection (ANSWERED)
Embedding similarity, k-NN example banks, MMR diversity, and eval-driven example curation for few-shot prompts.
TL;DR — Quick Answer
Store labeled examples in a vector index and retrieve the top-k most similar to the current query, optionally diversifying with MMR to avoid near-duplicates. Refresh the example bank from production failures and eval gaps, and cap total shots to control latency and token cost.
The Interview Question
How do you dynamically select few-shot examples at runtime instead of using a static set?
Deep Explanation
Why dynamic selection
Sign in to unlock full answer
Get deep explanations, PDF export & all Prompt Engineering questions
- 21 more sections of deep explanation
- Real-world examples
- Common mistakes
- Interviewer expectations
- Follow-up questions
Few-ShotDynamic ExamplesEmbeddingsRAGOpenAIPinecone