AI & LLM integration patterns
Call LLM APIs, structure prompts, parse responses, and ship AI features safely.
How to Filter Blocked Words in Python
Scans input text against a moderation blocklist, returning blocked terms and their counts.
MODERATION_BLOCKLIST = {"spam", "scam", "fraud", "phishing", "malware", "abuse"}
def scan_text(text: str) -> dict:
normalized = text.lower()
words = normalized.replace(".", " ").replace(",", " ").replace("!", " ").replace("?", " ").split()
found_terms = []
for word in words:
if word in MO…How to Filter Toxic Keywords in Python
Filter toxic keywords from text by replacing each occurrence with asterisks, useful as a basic guardrail for LLM inputs.
TOXIC_KEYWORDS = ["insult", "threat", "hate", "violence", "spam"]
def guardrails_filter(text: str, keywords: list[str] | None = None) -> str:
"""Filter out toxic keywords from the given text.
Args:
text: The input text to filter.
keywords: Optional keyword list. Defaults to TOXIC_KEYWORDS.
…How to Keep Last K Turns in a Memory Buffer in Python
A TurnBuffer class using deque with maxlen to keep only the most recent k conversation turns in memory for LLM context.
from collections import deque
class TurnBuffer:
def __init__(self, k):
self.k = k
self.turns = deque(maxlen=k)
def add(self, turn):
self.turns.append(turn)
def last_k(self):
return list(self.turns)
if __name__ == "__main__":
buffer = TurnBuffer(3)
buffer.add("tu…How to Log Prompts and Completions as JSONL Audit Files in Python
Read a JSONL file of LLM prompt–completion pairs, compute totals and averages, then write an audit summary with timestamps.
import json
from pathlib import Path
from datetime import datetime
def audit_jsonl(filepath):
logs = []
with open(filepath, encoding="utf-8") as f:
for line in f:
line = line.strip()
if not line:
continue
entry = json.loads(line)
logs.ap…How to Mock OpenAI Tool Call Messages in Python
Create an assistant message with a function tool call in OpenAI's chat format, useful for testing and mocking.
from openai import OpenAI
def mock_tool_call(tool_name: str, arguments: dict) -> dict:
"""Simulate a tool call message in OpenAI style."""
return {
"role": "assistant",
"content": None,
"tool_calls": [
{
"id": "call_" + "a1b2c3d4e5f6",
"type…How to Mock an LLM Client in Python
Create a simple mock LLM client that returns a canned completion for testing or development without a real API.
from dataclasses import dataclass
@dataclass
class MockLLMClient:
canned_response: str = "This is a canned completion."
def complete(self, prompt: str) -> str:
return f"{self.canned_response} [to: {prompt[:20]}]"
if __name__ == "__main__":
client = MockLLMClient()
result = client.complete(…How to Parse Chat Completion JSON in Python
Parse a mock OpenAI chat completion JSON response into a clean dictionary with content, finish reason, and model.
import json
def parse_chat_response(raw: str) -> dict:
data = json.loads(raw)
choice = data["choices"][0]
return {
"content": choice["message"]["content"],
"finish_reason": choice["finish_reason"],
"model": data["model"],
}
if __name__ == "__main__":
mock_response = '''
…How to Parse JSON from LLM Model Output Fence in Python
Extract and parse a JSON object from a language model's output that may be wrapped in triple-backtick fences with an optional language tag.
import json
import re
def parse_json_from_fence(text):
"""
Extract JSON object from a model output that may be wrapped in
triple-backtick fences with optional language tag.
"""
# Match content insideHow to Parse an LLM Response in Python
This code parses a JSON string from an LLM response, stripping code fences and handling common issues like whitespace, returning a Python dictionary.
import json
from typing import Any, Dict, List
def parse_llm_response(response: str) -> Dict[str, Any]:
"""Parse a JSON string from an LLM response, handling common edge cases."""
# Remove code fences if present
cleaned = response.strip()
if cleaned.startswith("How to Redact Emails and Phones Before Sending to an LLM in Python
This code uses regular expressions to replace email addresses and US phone numbers with [EMAIL] and [PHONE] placeholders before any LLM processing.
import re
def redact_pii(text: str) -> str:
# Replace email addresses with [EMAIL]
text = re.sub(r'[\w.+-]+@[\w-]+\.[\w.-]+', '[EMAIL]', text)
# Replace phone numbers (US format) with [PHONE]
text = re.sub(r'\(?\d{3}\)?[-.\s]?\d{3}[-.\s]?\d{4}', '[PHONE]', text)
return text
if __name__ == "__main…How to Render a Jinja-like Template from a Dict in Python
Replace {{placeholders}} in a string using values from a Python dict with a simple regex-based template renderer.
import re
def render_template(template, context):
pattern = re.compile(r"\{\{\s*(\w+)\s*\}\}")
def replace(match):
key = match.group(1)
return str(context.get(key, ""))
return pattern.sub(replace, template)
if __name__ == "__main__":
template = "Hello {{name}}, you have {{count}} new …How to Repair Malformed JSON Braces Heuristically in Python
Heuristically fix malformed JSON by balancing braces and quotes, using a stack-based approach to add missing closing characters.
import json
import re
def repair_json(text: str) -> str:
"""Heuristically repair malformed JSON by balancing braces and quotes."""
# Trim whitespace and handle leading/trailing garbage
text = text.strip()
# Remove common non-JSON decorations
text = re.sub(r'^(How to Retry LLM Calls on Rate Limit Errors in Python
Implement a retry mechanism with exponential backoff for LLM API calls that raises a custom RateLimitError, using a mock function to demonstrate the pattern.
import time
import random
def mock_llm_call():
"""Simulates an LLM API call that may raise a rate limit error."""
if random.random() < 0.4: # 40% chance of rate limit
raise RateLimitError("Rate limit exceeded. Try again later.")
return {"response": "Hello world from mock LLM"}
class RateLimitE…How to Serialize Chat Messages to a JSON File in Python
Writes a list of chat message dicts to a JSON file with metadata like export time and message count.
import json
from pathlib import Path
from datetime import datetime
def serialize_messages(messages, output_path):
data = {
"exported_at": datetime.now().isoformat(),
"count": len(messages),
"messages": messages
}
Path(output_path).write_text(
json.dumps(data, indent=2, ensu…How to Stream Tokens from a Mock LLM in Python
Simulate real-time LLM streaming by yielding tokens one at a time with a delay, making it easy to test streaming UIs.
import time
from typing import Generator
def stream_tokens(text: str, delay: float = 0.05) -> Generator[str, None, None]:
"""Simulate an LLM streaming tokens word by word."""
for word in text.split():
yield word
time.sleep(delay)
if __name__ == "__main__":
sample = "Hello world! This is…How to Summarize Old Conversation Turns in Python
Compress old conversation turns into a brief summary while keeping recent turns intact for LLM context management.
from datetime import datetime, timedelta
def summarize_old_turns(conversation, max_turns=5):
"""Compress turns older than max_turns into a brief summary."""
if len(conversation) <= max_turns:
return conversation, ""
old_turns = conversation[:-max_turns]
recent_turns = conversation[-max_turns…How to Truncate Text to a Token Budget in Python
Truncate a string to a maximum token budget for LLM context using the tiktoken library and OpenAI's tokenizer.
import tiktoken
def truncate_to_token_budget(text, max_tokens, model="gpt-3.5-turbo"):
enc = tiktoken.encoding_for_model(model)
tokens = enc.encode(text)
if len(tokens) <= max_tokens:
return text
truncated_tokens = tokens[:max_tokens]
return enc.decode(truncated_tokens)
if __name__ == "__…How to Validate JSON Output Against a Dict Schema in Python
Validate JSON-like data against a simple dict schema with type checking and descriptive error messages using only the Python standard library.
from typing import Dict, Any, List, Union
def validate_json(data: Any, schema: Dict[str, str]) -> List[str]:
"""
Validate JSON-like data against a simple dict schema.
Schema format: {field_name: expected_type} where type is one of:
'str', 'int', 'float', 'bool', 'list', 'dict', 'any'
Returns list …How to Validate LLM Output in Python
A beginner-friendly DataValidator class that checks required fields and type constraints on LLM-generated or user JSON data.
import json
from typing import Any, Dict, List, Optional
class DataValidator:
"""Simple helper for validating LLM-generated or user data."""
def __init__(self, required_fields: List[str], schema: Optional[Dict[str, str]] = None):
self.required_fields = required_fields
self.schema = schema or…How to build a function calling schema dict in Python
Build an OpenAI-compatible function calling schema dictionary with a helper function that takes name, description, parameters, and required fields.
import json
from typing import Dict, Any, List, Optional
def build_function_schema(
name: str,
description: str,
parameters: Optional[Dict[str, Any]] = None,
required: Optional[List[str]] = None
) -> Dict[str, Any]:
"""Build an OpenAI-compatible function calling schema dictionary."""
schema: …How to build a mock RAG pipeline in Python
Build a minimal Retrieval-Augmented Generation pipeline that retrieves the best-matching document by keyword overlap and generates a template-based answer.
def simple_rag_pipeline(question, documents):
"""
A minimal mock RAG pipeline: retrieve relevant context, then generate an answer.
"""
# Step 1: Retrieve — mock retrieval by simple keyword scoring
scores = []
for doc in documents:
doc_words = set(doc.lower().split())
question_wo…How to cache embeddings with a Python dict to avoid recomputation
Caches embeddings computed from text in a dictionary keyed by SHA-256 hash, returning cached results for repeated calls.
import hashlib
import time
class EmbeddingCache:
def __init__(self):
self.cache = {}
def _hash_text(self, text):
return hashlib.sha256(text.encode()).hexdigest()
def get_embedding(self, text, compute_func):
key = self._hash_text(text)
if key not in self.cache:
…How to compute ROUGE recall in Python
Compute ROUGE recall by counting token overlap between a reference and candidate summary with pure Python.
def rouge_recall(reference, candidate):
ref_tokens = reference.lower().split()
cand_tokens = candidate.lower().split()
ref_counts = {}
for token in ref_tokens:
ref_counts[token] = ref_counts.get(token, 0) + 1
cand_counts = {}
for token in cand_tokens:
cand_counts[token] = cand…How to compute exact match metric in Python
Computes the exact match (EM) metric for LLM outputs by normalizing text and comparing predictions against references.
def compute_exact_match(predictions, references):
def normalize(text):
import re
text = text.lower().strip()
text = re.sub(r'\b(a|an|the)\b', ' ', text)
text = re.sub(r'[^a-z0-9\s]', '', text)
text = ' '.join(text.split())
return text
matches = sum(1 for pred, r…Browse by section
Each section groups closely related Python snippets.
AI & LLM integration patterns — Python code examples
What you will find here
This page collects ai & llm integration patterns snippets — short, copy-ready Python you can paste into our free online IDE and run without installing anything. Each sample includes a plain-English explanation and the full source code.
Samples vs tutorials and challenges
Samples are quick reference — one concept per page. For step-by-step teaching, use our Python tutorials. To test yourself, try quizzes or coding challenges. Clean up style with the Python formatter.