Constrained Vector Index Design

U

@

·

Design vector index configuration under strict resource and performance constraints

78 copies0 forks
Design a vector index configuration within these strict constraints.

HARD CONSTRAINTS:
- Maximum memory: {{memory_limit_gb}}GB
- Query latency p99: {{latency_p99_ms}}ms
- Minimum recall: {{min_recall}}%
- Vector dimensions: {{dimensions}}
- Expected corpus size: {{corpus_size}} vectors

OUTPUT FORMAT:
```yaml
index_config:
  type: [HNSW|IVF|FLAT]
  parameters:
    # Index-specific params
  memory_estimate_gb: <number>
  expected_latency_p99_ms: <number>
  expected_recall: <number>
  
justification:
  type_selection: <reason>
  parameter_choices: <reason>
  tradeoffs: <list>
```

Configuration must satisfy ALL constraints. Explain any tradeoffs made.

Details

Category

Coding

Use Cases

Index configurationResource optimizationPerformance tuning

Works Best With

claude-sonnet-4-20250514gpt-4o
Created Shared

Create your own prompt vault and start sharing