"""
Pattern 038: Prompt Routing
Description:
Prompt Routing intelligently routes queries to specialized prompts or models
based on query characteristics such as type, complexity, domain, or intent.
This enables efficient resource utilization and improved response quality by
matching queries with the most appropriate handler.
Components:
- Router: Classifies and routes queries
- Specialized Prompts: Domain or task-specific prompts
- Routing Logic: Rules or model-based classification
- Fallback Handler: Default route for unmatched queries
- Performance Tracker: Monitors routing decisions
Use Cases:
- Multi-domain customer service
- Enterprise knowledge systems
- Cost optimization (route to cheaper models when possible)
- Specialized expert systems
- Content moderation and filtering
- Dynamic model selection
LangChain Implementation:
Uses classification chains and conditional routing to direct queries to
appropriate specialized prompts or models based on intent and complexity.
"""
import os
from typing import List, Dict, Any, Optional, Callable
from dataclasses import dataclass, field
from enum import Enum
from datetime import datetime
from dotenv import load_dotenv
from langchain_core.prompts import ChatPromptTemplate
from langchain_core.output_parsers import StrOutputParser
from langchain_openai import ChatOpenAI
load_dotenv()
class QueryType(Enum):
"""Types of queries for routing."""
TECHNICAL = "technical"
CREATIVE = "creative"
ANALYTICAL = "analytical"
CONVERSATIONAL = "conversational"
FACTUAL = "factual"
CODE = "code"
MATH = "math"
GENERAL = "general"
class QueryComplexity(Enum):
"""Complexity levels for queries."""
SIMPLE = "simple"
MODERATE = "moderate"
COMPLEX = "complex"
EXPERT = "expert"
@dataclass
class Route:
"""A routing destination."""
name: str
query_types: List[QueryType]
prompt: ChatPromptTemplate
model: str = "gpt-3.5-turbo"
temperature: float = 0.7
description: str = ""
priority: int = 1 # Higher priority routes checked first
@dataclass
class RoutingDecision:
"""Result of routing decision."""
query: str
selected_route: str
query_type: QueryType
complexity: QueryComplexity
confidence: float
reasoning: str
timestamp: datetime = field(default_factory=datetime.now)
@dataclass
class RoutedResponse:
"""Response from routed query."""
query: str
response: str
route_used: str
routing_decision: RoutingDecision
execution_time: float
timestamp: datetime = field(default_factory=datetime.now)
class QueryRouter:
"""
Routes queries to specialized prompts based on classification.
Features:
- Automatic query classification
- Rule-based and model-based routing
- Multiple routing strategies
- Performance tracking
- Fallback handling
"""
def __init__(
self,
classification_model: str = "gpt-3.5-turbo",
use_llm_classification: bool = True
):
self.routes: Dict[str, Route] = {}
self.default_route: Optional[Route] = None
self.use_llm_classification = use_llm_classification
if use_llm_classification:
self.classifier_llm = ChatOpenAI(
model=classification_model,
temperature=0.1 # Low temperature for consistent classification
)
# Classification prompt
self.classification_prompt = ChatPromptTemplate.from_messages([
("system", """You are a query classifier. Analyze the query and classify it.
Available Types:
- TECHNICAL: Software, hardware, technical troubleshooting
- CREATIVE: Writing, art, brainstorming, creative ideas
- ANALYTICAL: Data analysis, research, critical thinking
- CONVERSATIONAL: Casual chat, greetings, small talk
- FACTUAL: Facts, definitions, information lookup
- CODE: Programming, code generation, debugging
- MATH: Mathematics, calculations, formulas
- GENERAL: General questions that don't fit other categories
Complexity Levels:
- SIMPLE: Straightforward, single-step questions
- MODERATE: Multi-step but manageable questions
- COMPLEX: Requires deep reasoning or multiple components
- EXPERT: Highly specialized or advanced questions
Respond in this format:
TYPE: [query type]
COMPLEXITY: [complexity level]
CONFIDENCE: [0.0-1.0]
REASONING: [brief explanation]"""),
("user", "Query: {query}\n\nClassify this query.")
])
self.routing_history: List[RoutingDecision] = []
def add_route(self, route: Route):
"""Add a routing destination."""
self.routes[route.name] = route
def set_default_route(self, route: Route):
"""Set the default fallback route."""
self.default_route = route
def classify_query(self, query: str) -> tuple[QueryType, QueryComplexity, float, str]:
"""
Classify query using LLM or rules.
Returns:
Tuple of (query_type, complexity, confidence, reasoning)
"""
if self.use_llm_classification and hasattr(self, 'classifier_llm'):
# LLM-based classification
chain = self.classification_prompt | self.classifier_llm | StrOutputParser()
classification = chain.invoke({"query": query})
# Parse classification result
query_type, complexity, confidence, reasoning = self._parse_classification(classification)
else:
# Rule-based classification
query_type, complexity, confidence, reasoning = self._rule_based_classification(query)
return query_type, complexity, confidence, reasoning
def _parse_classification(self, classification: str) -> tuple[QueryType, QueryComplexity, float, str]:
"""Parse LLM classification output."""
query_type = QueryType.GENERAL
complexity = QueryComplexity.MODERATE
confidence = 0.5
reasoning = ""
lines = classification.split('\n')
for line in lines:
if line.startswith("TYPE:"):
type_str = line.split(':')[1].strip().upper()
try:
query_type = QueryType[type_str]
except (KeyError, ValueError):
pass
elif line.startswith("COMPLEXITY:"):
complexity_str = line.split(':')[1].strip().upper()
try:
complexity = QueryComplexity[complexity_str]
except (KeyError, ValueError):
pass
elif line.startswith("CONFIDENCE:"):
try:
confidence = float(line.split(':')[1].strip())
except (ValueError, IndexError):
pass
elif line.startswith("REASONING:"):
reasoning = line.split(':', 1)[1].strip()
return query_type, complexity, confidence, reasoning
def _rule_based_classification(self, query: str) -> tuple[QueryType, QueryComplexity, float, str]:
"""Simple rule-based classification."""
query_lower = query.lower()
# Classify by keywords
if any(word in query_lower for word in ['code', 'program', 'function', 'class', 'debug']):
query_type = QueryType.CODE
reasoning = "Keywords suggest coding question"
elif any(word in query_lower for word in ['calculate', 'solve', 'equation', 'math']):
query_type = QueryType.MATH
reasoning = "Keywords suggest mathematical question"
elif any(word in query_lower for word in ['write', 'create', 'imagine', 'story', 'poem']):
query_type = QueryType.CREATIVE
reasoning = "Keywords suggest creative task"
elif any(word in query_lower for word in ['analyze', 'compare', 'evaluate', 'research']):
query_type = QueryType.ANALYTICAL
reasoning = "Keywords suggest analytical task"
elif any(word in query_lower for word in ['hello', 'hi', 'how are you', 'thanks']):
query_type = QueryType.CONVERSATIONAL
reasoning = "Conversational greeting or response"
elif any(word in query_lower for word in ['what is', 'define', 'who is', 'when']):
query_type = QueryType.FACTUAL
reasoning = "Factual information request"
else:
query_type = QueryType.GENERAL
reasoning = "General question"
# Estimate complexity by length and question marks
word_count = len(query.split())
question_marks = query.count('?')
if word_count < 10 and question_marks <= 1:
complexity = QueryComplexity.SIMPLE
elif word_count < 30:
complexity = QueryComplexity.MODERATE
elif word_count RoutingDecision:
"""
Determine the best route for a query.
Args:
query: The query to route
Returns:
RoutingDecision with selected route
"""
# Classify query
query_type, complexity, confidence, reasoning = self.classify_query(query)
# Find matching route
selected_route = None
# Sort routes by priority
sorted_routes = sorted(self.routes.values(), key=lambda r: r.priority, reverse=True)
for route in sorted_routes:
if query_type in route.query_types:
selected_route = route.name
break
# Use default route if no match
if not selected_route and self.default_route:
selected_route = self.default_route.name
decision = RoutingDecision(
query=query,
selected_route=selected_route or "none",
query_type=query_type,
complexity=complexity,
confidence=confidence,
reasoning=reasoning
)
self.routing_history.append(decision)
return decision
def execute_route(self, query: str) -> RoutedResponse:
"""
Route and execute query.
Args:
query: The query to process
Returns:
RoutedResponse with result
"""
start_time = datetime.now()
# Get routing decision
decision = self.route(query)
# Get the selected route
if decision.selected_route not in self.routes:
if self.default_route:
route = self.default_route
else:
return RoutedResponse(
query=query,
response="No appropriate route found for this query.",
route_used="none",
routing_decision=decision,
execution_time=0.0
)
else:
route = self.routes[decision.selected_route]
# Execute using the selected route
llm = ChatOpenAI(model=route.model, temperature=route.temperature)
chain = route.prompt | llm | StrOutputParser()
response = chain.invoke({"query": query})
end_time = datetime.now()
execution_time = (end_time - start_time).total_seconds()
return RoutedResponse(
query=query,
response=response,
route_used=route.name,
routing_decision=decision,
execution_time=execution_time
)
def get_routing_stats(self) -> Dict[str, Any]:
"""Get statistics about routing decisions."""
if not self.routing_history:
return {"total_routes": 0}
by_type = {}
by_complexity = {}
by_route = {}
for decision in self.routing_history:
by_type[decision.query_type.value] = by_type.get(decision.query_type.value, 0) + 1
by_complexity[decision.complexity.value] = by_complexity.get(decision.complexity.value, 0) + 1
by_route[decision.selected_route] = by_route.get(decision.selected_route, 0) + 1
avg_confidence = sum(d.confidence for d in self.routing_history) / len(self.routing_history)
return {
"total_routes": len(self.routing_history),
"by_type": by_type,
"by_complexity": by_complexity,
"by_route": by_route,
"average_confidence": avg_confidence
}
def demonstrate_prompt_routing():
"""
Demonstrates prompt routing with specialized handlers for different query types.
"""
print("=" * 80)
print("PROMPT ROUTING DEMONSTRATION")
print("=" * 80)
# Create router
router = QueryRouter(use_llm_classification=True)
# Define specialized routes
# 1. Code route
code_prompt = ChatPromptTemplate.from_messages([
("system", "You are an expert programmer. Provide clear, well-commented code solutions."),
("user", "{query}")
])
code_route = Route(
name="code_expert",
query_types=[QueryType.CODE],
prompt=code_prompt,
temperature=0.3,
description="Handles programming and code questions",
priority=10
)
router.add_route(code_route)
# 2. Creative route
creative_prompt = ChatPromptTemplate.from_messages([
("system", "You are a creative writer. Provide imaginative, engaging responses."),
("user", "{query}")
])
creative_route = Route(
name="creative_writer",
query_types=[QueryType.CREATIVE],
prompt=creative_prompt,
temperature=0.9,
description="Handles creative writing tasks",
priority=9
)
router.add_route(creative_route)
# 3. Math route
math_prompt = ChatPromptTemplate.from_messages([
("system", "You are a mathematics expert. Provide step-by-step solutions with clear explanations."),
("user", "{query}")
])
math_route = Route(
name="math_expert",
query_types=[QueryType.MATH],
prompt=math_prompt,
temperature=0.1,
description="Handles mathematical questions",
priority=10
)
router.add_route(math_route)
# 4. Analytical route
analytical_prompt = ChatPromptTemplate.from_messages([
("system", "You are an analytical expert. Provide thorough analysis with evidence and reasoning."),
("user", "{query}")
])
analytical_route = Route(
name="analytical_expert",
query_types=[QueryType.ANALYTICAL],
prompt=analytical_prompt,
temperature=0.5,
description="Handles analytical and research questions",
priority=8
)
router.add_route(analytical_route)
# 5. Default route
default_prompt = ChatPromptTemplate.from_messages([
("system", "You are a helpful AI assistant."),
("user", "{query}")
])
default_route = Route(
name="general_assistant",
query_types=[QueryType.GENERAL, QueryType.CONVERSATIONAL, QueryType.FACTUAL, QueryType.TECHNICAL],
prompt=default_prompt,
temperature=0.7,
description="Handles general questions",
priority=1
)
router.add_route(default_route)
router.set_default_route(default_route)
# Show available routes
print("\n" + "=" * 80)
print("Available Routes")
print("=" * 80)
for name, route in router.routes.items():
print(f"\n{name}:")
print(f" Description: {route.description}")
print(f" Query Types: {[qt.value for qt in route.query_types]}")
print(f" Temperature: {route.temperature}")
print(f" Priority: {route.priority}")
# Test different query types
test_queries = [
"Write a Python function to calculate fibonacci numbers",
"Write a short poem about artificial intelligence",
"What is 156 multiplied by 327?",
"Analyze the pros and cons of renewable energy",
"Hello! How are you today?",
]
print("\n" + "=" * 80)
print("Routing Queries")
print("=" * 80)
for i, query in enumerate(test_queries, 1):
print(f"\n{'-' * 80}")
print(f"Query {i}: {query}")
print(f"{'-' * 80}")
result = router.execute_route(query)
print(f"\n[Routing Decision]")
print(f" Route: {result.route_used}")
print(f" Type: {result.routing_decision.query_type.value}")
print(f" Complexity: {result.routing_decision.complexity.value}")
print(f" Confidence: {result.routing_decision.confidence:.2f}")
print(f" Reasoning: {result.routing_decision.reasoning}")
print(f"\n[Response]")
response_preview = result.response[:200] + "..." if len(result.response) > 200 else result.response
print(f" {response_preview}")
print(f"\n Execution Time: {result.execution_time:.2f}s")
# Show routing statistics
print("\n" + "=" * 80)
print("Routing Statistics")
print("=" * 80)
stats = router.get_routing_stats()
print(f"\nTotal Queries Routed: {stats['total_routes']}")
print(f"Average Confidence: {stats['average_confidence']:.2f}")
print("\nQueries by Type:")
for qtype, count in stats['by_type'].items():
print(f" - {qtype}: {count}")
print("\nQueries by Complexity:")
for complexity, count in stats['by_complexity'].items():
print(f" - {complexity}: {count}")
print("\nQueries by Route:")
for route, count in stats['by_route'].items():
print(f" - {route}: {count}")
# Summary
print("\n" + "=" * 80)
print("SUMMARY")
print("=" * 80)
print("""
Prompt Routing provides:
✓ Intelligent query classification
✓ Specialized prompt selection
✓ Model and parameter optimization
✓ Cost-effective resource allocation
✓ Domain expertise routing
✓ Performance tracking
This pattern excels at:
- Multi-domain systems
- Cost optimization
- Quality through specialization
- Dynamic model selection
- Enterprise knowledge systems
- Efficient resource use
Routing components:
1. Router: Classifies and routes queries
2. Specialized Prompts: Domain-specific handlers
3. Classification: LLM or rule-based
4. Fallback: Default handler
5. Tracker: Performance monitoring
Query classification:
- Query Type: Domain/category (code, creative, math, etc.)
- Complexity: Difficulty level (simple to expert)
- Confidence: Classification certainty
- Reasoning: Why this classification
Routing strategies:
- Rule-based: Keyword matching, patterns
- Model-based: LLM classification
- Hybrid: Combine rules and models
- Priority-based: Check high-priority routes first
- Confidence-based: Threshold for routing
Route characteristics:
- Name: Identifier
- Query Types: What it handles
- Prompt: Specialized prompt template
- Model: Which LLM to use
- Temperature: Generation parameters
- Priority: Routing precedence
Benefits:
- Specialization: Expert prompts for domains
- Efficiency: Right model for right task
- Cost: Use cheaper models when appropriate
- Quality: Optimized prompts and parameters
- Scalability: Easy to add new routes
- Monitoring: Track routing decisions
Use Prompt Routing when you need:
- Multi-domain question answering
- Cost-optimized model selection
- Specialized expert systems
- Quality through domain expertise
- Dynamic resource allocation
- Large-scale production systems
""")
if __name__ == "__main__":
demonstrate_prompt_routing()
Agentic Design Pattern: Prompt Routing
Pattern 038: Prompt Routing – Line-by-Line Explanation
Overview
This document provides a detailed line-by-line explanation of the Prompt Routing pattern implementation.
Pattern Description
Pattern 038: Prompt Routing
Description: Prompt Routing intelligently routes queries to specialized prompts or models based on query characteristics such as type, complexity, domain, or intent. This enables efficient resource utilization and improved response quality by matching queries with the most appropriate handler.
Components: – Router: Classifies and routes queries – Specialized Prompts: Domain or task-specific prompts – Routing Logic: Rules or model-based classification – Fallback Handler: Default route for unmatched queries – Performance Tracker: Monitors routing decisions
Use Cases: – Multi-domain customer service – Enterprise knowledge systems – Cost optimization (route to cheaper models when possible) – Specialized expert systems – Content moderation and filtering – Dynamic model selection
LangChain Implementation: Uses classification chains and conditional routing to direct queries to appropriate specialized prompts or models based on intent and complexity.
File Statistics
- Total Lines: 595
- Classes: 6
- Functions: 1
- Constants: 0
Module Documentation
"""
Pattern 038: Prompt Routing
Description:
Prompt Routing intelligently routes queries to specialized prompts or models
based on query characteristics such as type, complexity, domain, or intent.
This enables efficient resource utilization and improved response quality by
matching queries with the most appropriate handler.
Components:
- Router: Classifies and routes queries
- Specialized Prompts: Domain or task-specific prompts
- Routing Logic: Rules or model-based classification
- Fallback Handler: Default route for unmatched queries
- Performance Tracker: Monitors routing decisions
Use Cases:
- Multi-domain customer service
- Enterprise knowledge systems
- Cost optimization (route to cheaper models when possible)
- Specialized expert systems
- Content moderation and filtering
- Dynamic model selection
LangChain Implementation:
Uses classification chains and conditional routing to direct queries to
appropriate specialized prompts or models based on intent and complexity.
"""
Purpose: This module implements the Prompt Routing pattern, providing a structured approach to agent design and implementation.
Import Statements
Standard Library Imports
import dataclasses.dataclass
import dataclasses.field
import datetime.datetime
import enum.Enum
import langchain_core.output_parsers.StrOutputParser
import langchain_core.prompts.ChatPromptTemplate
import os
import typing.Any
import typing.Callable
import typing.Dict
import typing.List
import typing.Optional
Purpose: Core Python functionality:
– dataclasses.dataclass: Simplified class definitions with @dataclass
– dataclasses.field: Simplified class definitions with @dataclass
– datetime.datetime: Date and time handling
– enum.Enum: Enumeration support
– langchain_core.output_parsers.StrOutputParser: Regular expression operations
– langchain_core.prompts.ChatPromptTemplate: Regular expression operations
– os: Environment variable access and system operations
– typing.Any: Type hints for better code documentation (List, Dict, Any, Optional)
– typing.Callable: Type hints for better code documentation (List, Dict, Any, Optional)
– typing.Dict: Type hints for better code documentation (List, Dict, Any, Optional)
– typing.List: Type hints for better code documentation (List, Dict, Any, Optional)
– typing.Optional: Type hints for better code documentation (List, Dict, Any, Optional)
LangChain Imports
from langchain_openai.ChatOpenAI
Purpose: LangChain framework components: – LLM Interface: Wrapper for OpenAI models – Chains: Composable building blocks for complex workflows
Other Imports
from dotenv.load_dotenv
Purpose: Additional dependencies for specialized functionality.
Classes
Class: QueryType
Lines 44-53
Inherits from: Enum
Purpose:
Types of queries for routing.
Class: QueryComplexity
Lines 56-61
Inherits from: Enum
Purpose:
Complexity levels for queries.
Class: Route
Lines 65-73
Purpose:
A routing destination.
Class: RoutingDecision
Lines 77-85
Purpose:
Result of routing decision.
Class: RoutedResponse
Lines 89-96
Purpose:
Response from routed query.
Class: QueryRouter
Lines 99-369
Purpose:
Routes queries to specialized prompts based on classification.
Features:
- Automatic query classification
- Rule-based and model-based routing
- Multiple routing strategies
- Performance tracking
- Fallback handling
Methods:
__init__(self, classification_model, use_llm_classification)
Lines 111-154
Purpose: Initialize the class instance with required parameters.
Parameters:
– classification_model: Language model instance
– use_llm_classification: Language model
add_route(self, route)
Lines 156-158
Purpose:
Add a routing destination.
Parameters:
– route: Route
set_default_route(self, route)
Lines 160-162
Purpose:
Set the default fallback route.
Parameters:
– route: Route
classify_query(self, query)
Lines 164-182
Purpose:
Classify query using LLM or rules.
Returns:
Tuple of (query_type, complexity, confidence, reasoning)
Parameters:
– query: Input query or question
Returns: <ast.Subscript object at 0x000001A06D16EC90>
_parse_classification(self, classification)
Lines 184-213
Purpose:
Parse LLM classification output.
Parameters:
– classification: Classification
Returns: <ast.Subscript object at 0x000001A06D178390>
_rule_based_classification(self, query)
Lines 215-257
Purpose:
Simple rule-based classification.
Parameters:
– query: Input query or question
Returns: <ast.Subscript object at 0x000001A06D6AA390>
route(self, query)
Lines 259-298
Purpose:
Determine the best route for a query.
Args:
query: The query to route
Returns:
RoutingDecision with selected route
Parameters:
– query: Input query or question
Returns: RoutingDecision
execute_route(self, query)
Lines 300-345
Purpose:
Route and execute query.
Args:
query: The query to process
Returns:
RoutedResponse with result
Parameters:
– query: Input query or question
Returns: RoutedResponse
get_routing_stats(self)
Lines 347-369
Purpose:
Get statistics about routing decisions.
Parameters:
Returns: <ast.Subscript object at 0x000001A06D668D50>
Functions
demonstrate_prompt_routing()
Lines 372-590
Purpose:
Demonstrates prompt routing with specialized handlers for different query types.
Key Concepts Summary
Pattern Overview
The Prompt Routing pattern provides a structured approach to building AI agents with specific capabilities and behaviors. This implementation leverages LangChain’s composable components to create a robust and maintainable solution.
Core Components
- Agent Class: Main implementation of the pattern logic
- Prompt Engineering: Carefully crafted prompts to guide LLM behavior
- Tool Integration: Optional tools for extended functionality
- Memory Management: State handling across interactions
- Output Parsing: Structured response processing
When to Use
✅ Good for: – Scenarios requiring this specific pattern’s strengths – Applications needing structured agent behavior – Systems where explainability is important – Production environments requiring reliability
❌ Not ideal for: – Simple single-prompt tasks – Scenarios requiring different pattern characteristics – Resource-constrained environments (if computationally intensive)
Best Practices Demonstrated
1. Environment Configuration
- ✅ Use
.envfiles for API keys - ✅ Never hardcode credentials
- ✅ Validate environment variables on startup
2. Error Handling
- ✅ Graceful degradation on failures
- ✅ Informative error messages
- ✅ Logging for debugging
3. Prompt Engineering
- ✅ Clear, specific instructions
- ✅ Structured output formats
- ✅ Few-shot examples when beneficial
4. Code Organization
- ✅ Separation of concerns (class-based design)
- ✅ Type hints for clarity
- ✅ Docstrings for documentation
5. Performance Optimization
- ✅ Appropriate temperature settings
- ✅ Token budget management
- ✅ Caching where applicable
LangChain Components Used
- ChatOpenAI: OpenAI model interface
- Chains: Sequential component composition
Why LangChain? – Composable abstractions for LLM applications – Rich ecosystem of integrations – Production-ready patterns – Active community and development
Execution Flow
1. Initialize agent with configuration
↓
2. Load environment and dependencies
↓
3. Prepare prompt with input data
↓
4. Invoke LLM through LangChain
↓
5. Parse and validate response
↓
6. Return formatted result
Key Decision Points: – Model selection (temperature, model type) – Prompt structure (few-shot vs zero-shot) – Error handling strategy – Output format validation
Performance Considerations
Computational Cost
- LLM Calls: Primary cost factor (based on tokens)
- Latency: Depends on model and input size
- Memory: Minimal for most patterns
Optimization Strategies
- Prompt Optimization: Minimize tokens while maintaining clarity
- Caching: Cache repeated queries/responses
- Model Selection: Use smaller models when appropriate
- Batch Processing: Group similar requests
- Streaming: For long responses, stream tokens
Cost Estimates (Approximate)
- GPT-3.5-turbo: ~$0.002 per request
- GPT-4: ~$0.03 per request
- GPT-4-turbo: ~$0.01 per request
Actual costs vary based on input/output token counts
Common Pitfalls to Avoid
1. Prompt Design Issues
❌ Vague or ambiguous instructions
✅ Clear, specific, testable prompts
2. Error Handling
❌ Ignoring API errors or rate limits
✅ Implement retry logic and graceful degradation
3. Token Management
❌ Exceeding model token limits
✅ Truncate inputs, chunk processing
4. Security
❌ Exposing API keys in code
✅ Use environment variables
5. Testing
❌ Only testing happy paths
✅ Test edge cases, failures, malformed inputs
6. Cost Management
❌ Unmonitored API usage
✅ Implement budgets, alerts, and monitoring
Further Improvements
Production Enhancements
- Monitoring: Add logging, metrics, tracing
- Validation: Input/output schema validation
- Rate Limiting: Implement request throttling
- Caching: Redis or in-memory cache for responses
- Async Support: Non-blocking operations for scalability
Feature Extensions
- Multi-Model Support: Switch between providers
- Fallback Models: Graceful degradation strategy
- Custom Tools: Domain-specific function calling
- Advanced Memory: Vector stores for long-term memory
- Evaluation: Automated testing and benchmarking
Code Quality
- Testing: Unit tests, integration tests
- Documentation: API docs, usage examples
- Type Safety: Stricter type hints
- Configuration: Externalized config files
Related Patterns
- Pattern 037
- Pattern 039
Pattern Families: – Reasoning Patterns (001-011) – Multi-Agent Patterns (012-019) – Tool Use Patterns (020-025) – Memory Patterns (026-031) – And many more…
Summary
The Prompt Routing pattern provides a powerful approach to building AI agents with specific capabilities. This implementation demonstrates:
✅ Key Strengths: – Clean, maintainable code structure – LangChain integration for composability – Type hints and documentation – Production-ready patterns
🎯 Best Use Cases: – Applications requiring this pattern’s characteristics – Systems needing structured agent behavior – Production environments requiring reliability
📚 Learning Resources: – LangChain Documentation: https://python.langchain.com/ – Pattern Documentation: See related patterns – OpenAI API Reference: https://platform.openai.com/docs/
File: 038_prompt_routing.py
Total Lines: 595
Pattern Category: AI Agent Design Pattern #038
Generated from actual implementation code
For educational and reference purposes