Few-Shot Graceful Degradation Implementation

Implement graceful degradation handlers based on these patterns.

Example 1 - Vector DB Timeout:
Scenario: Pinecone query exceeds 500ms timeout
Degradation: Fall back to keyword search with BM25
Code:
```python
async def query_with_fallback(query: str):
    try:
        return await asyncio.wait_for(vector_search(query), timeout=0.5)
    except asyncio.TimeoutError:
        logger.warning("Vector search timeout, using BM25 fallback")
        return keyword_search(query)
```

Example 2 - LLM Rate Limit:
Scenario: OpenAI returns 429 rate limit error
Degradation: Switch to smaller model with exponential backoff
Code:
```python
@retry(stop=stop_after_attempt(3), wait=wait_exponential(min=1, max=10))
async def generate_with_fallback(prompt: str):
    try:
        return await call_gpt4(prompt)
    except RateLimitError:
        return await call_gpt35(prompt)
```

Now implement degradation for:
Scenario: {{failure_scenario}}
Service: {{service_name}}
SLA Requirements: {{sla_requirements}}