Google Maps Agent

Build a location-aware AI agent that answers questions like 'Find the best rated sushi near Times Square' by combining Google Maps API with OneInfer's LLM via tool calling.

  • OneInfer Chat API
  • Google Maps API
  • Tool Calling
  • Python

Step-by-step guide

1

Define the Maps search tool

python
import openai
import requests
import json

client = openai.OpenAI(
    api_key="your-oneinfer-api-key",
    base_url="https://api.oneinfer.ai/v1"
)

MAPS_API_KEY = "your-google-maps-api-key"

tools = [
    {
        "type": "function",
        "function": {
            "name": "search_places",
            "description": "Search for places near a location using Google Maps.",
            "parameters": {
                "type": "object",
                "properties": {
                    "query": {"type": "string", "description": "Search query, e.g. 'sushi restaurants'"},
                    "location": {"type": "string", "description": "Address or landmark, e.g. 'Times Square, NYC'"},
                    "radius_meters": {"type": "integer", "description": "Search radius in meters (default 1000)"}
                },
                "required": ["query", "location"]
            }
        }
    }
]
2

Implement the tool execution

python
def search_places(query: str, location: str, radius_meters: int = 1000) -> str:
    # Geocode the location
    geo = requests.get(
        "https://maps.googleapis.com/maps/api/geocode/json",
        params={"address": location, "key": MAPS_API_KEY}
    ).json()

    loc = geo["results"][0]["geometry"]["location"]
    lat, lng = loc["lat"], loc["lng"]

    # Search for places
    places = requests.get(
        "https://maps.googleapis.com/maps/api/place/nearbysearch/json",
        params={
            "location": f"{lat},{lng}",
            "radius": radius_meters,
            "keyword": query,
            "key": MAPS_API_KEY,
            "rankby": "prominence"
        }
    ).json()

    results = places.get("results", [])[:5]
    formatted = []
    for p in results:
        formatted.append({
            "name": p.get("name"),
            "rating": p.get("rating"),
            "address": p.get("vicinity"),
            "open_now": p.get("opening_hours", {}).get("open_now")
        })
    return json.dumps(formatted, indent=2)
3

Run the agentic loop

python
def run_agent(user_query: str) -> str:
    messages = [
        {"role": "system", "content": "You are a helpful local guide. Use the search_places tool to answer location questions."},
        {"role": "user", "content": user_query}
    ]

    while True:
        response = client.chat.completions.create(
            model="meta-llama/Llama-3.3-70B-Instruct-Turbo",
            messages=messages,
            tools=tools,
            tool_choice="auto"
        )
        msg = response.choices[0].message

        if msg.tool_calls:
            messages.append(msg)
            for call in msg.tool_calls:
                args = json.loads(call.function.arguments)
                result = search_places(**args)
                messages.append({
                    "role": "tool",
                    "tool_call_id": call.id,
                    "content": result
                })
        else:
            return msg.content

print(run_agent("Find the top 3 rated sushi restaurants near Times Square that are open now"))