	In Browser
	StumbleUpon
	del.icio.us
	Google
	Google Buzz
	reddit
	LinkedIn

	Facebook
	Twitter
	Linkedin
	E-Mail

Generative AI > Google Gemini API > Accounts Receivable Chase Agent

Accounts Receivable Chase Agent

Author: Venkata Sudhakar

ShopMax India supplies electronics to corporate clients and government departments on credit terms. Some clients pay on day 30 as agreed; others slip to 60 or 90 days. Chasing overdue payments manually is time-consuming and awkward. An accounts receivable agent that tracks every outstanding invoice, grades each by overdue severity, and sends appropriately escalating messages keeps cash flowing without straining client relationships.

This tutorial builds an Accounts Receivable Chase Agent using ADK and Gemini. The agent scans open invoices, classifies each by days overdue, and generates a tiered message - gentle reminder at 7 days, firm notice at 30 days, formal demand at 60 days - with the exact amount and payment details included.

The below example shows the AR chase workflow for ShopMax India B2B clients.

from google.adk.agents import LlmAgent
from google.adk.tools import FunctionTool
from google.adk.runners import Runner
from google.adk.sessions import InMemorySessionService

# --- Outstanding B2B invoices ---
INVOICES = [
    {
        "invoice_id": "INV-B2B-1001",
        "client": "TechCorp Solutions Pvt Ltd",
        "contact": "Ramesh Gupta, CFO",
        "city": "Bangalore",
        "amount": 245000,
        "due_date": "2025-03-01",
        "days_overdue": 37,
        "payment_terms": "NET 30",
    },
    {
        "invoice_id": "INV-B2B-1002",
        "client": "Infosys BPO Hyderabad",
        "contact": "Sunita Rao, Accounts",
        "city": "Hyderabad",
        "amount": 128500,
        "due_date": "2025-03-15",
        "days_overdue": 23,
        "payment_terms": "NET 30",
    },
    {
        "invoice_id": "INV-B2B-1003",
        "client": "Chennai Municipal Corporation",
        "contact": "Deputy Commissioner Accounts",
        "city": "Chennai",
        "amount": 892000,
        "due_date": "2025-01-15",
        "days_overdue": 81,
        "payment_terms": "NET 45",
    },
]

# --- Tool: get overdue invoices ---
def get_overdue_invoices() -> dict:
    """Fetch all overdue B2B invoices sorted by severity.
    Returns:
        Categorised overdue invoices with severity tiers
    """
    critical = [i for i in INVOICES if i["days_overdue"] >= 60]
    overdue = [i for i in INVOICES if 30 <= i["days_overdue"] < 60]
    reminder = [i for i in INVOICES if i["days_overdue"] < 30]
    total_outstanding = sum(i["amount"] for i in INVOICES)
    return {
        "total_outstanding": total_outstanding,
        "critical_60_plus": critical,
        "overdue_30_to_59": overdue,
        "reminder_under_30": reminder,
    }

# --- Tool: get payment details for invoice ---
def get_payment_details(invoice_id: str) -> dict:
    """Get full invoice and payment details for a specific invoice.
    Args:
        invoice_id: Invoice identifier
    Returns:
        Invoice details with bank transfer instructions
    """
    invoice = next((i for i in INVOICES if i["invoice_id"] == invoice_id), None)
    if not invoice:
        return {"error": f"Invoice not found: {invoice_id}"}
    return {
        **invoice,
        "bank_name": "HDFC Bank",
        "account_name": "ShopMax India Pvt Ltd",
        "account_number": "50200012345678",
        "ifsc": "HDFC0001234",
        "upi": "shopmax@hdfcbank",
    }

# --- Agent setup ---
ar_tool = FunctionTool(func=get_overdue_invoices)
details_tool = FunctionTool(func=get_payment_details)

agent = LlmAgent(
    name="ar_chase_agent",
    model="gemini-2.0-flash",
    instruction="""You are an accounts receivable chase agent for ShopMax India.
1. Call get_overdue_invoices to get the full overdue summary.
2. Process invoices in severity order: critical (60+ days) first.
3. For each invoice, call get_payment_details and write an appropriate message:
   - Under 30 days: gentle reminder, friendly tone
   - 30-59 days: firm but professional, reference contract terms
   - 60+ days: formal demand letter tone, mention potential legal action if unresolved
4. Always include: invoice number, amount due, days overdue, payment instructions.
Format amounts as Rs with commas.""",
    tools=[ar_tool, details_tool],
)

session_service = InMemorySessionService()
runner = Runner(agent=agent, app_name="ar_chase", session_service=session_service)

import asyncio

async def run_chase(query: str) -> str:
    session = await session_service.create_session(app_name="ar_chase", user_id="finance")
    from google.genai.types import Content, Part
    msg = Content(role="user", parts=[Part(text=query)])
    result = ""
    async for event in runner.run_async(user_id="finance", session_id=session.id, new_message=msg):
        if event.is_final_response() and event.content:
            result = event.content.parts[0].text
    return result

response = asyncio.run(run_chase("Chase all overdue invoices with appropriate messages based on how long they are outstanding."))
print(response)

It gives the following output,

AR Chase Report - ShopMax India
Total outstanding: Rs 12,65,500  |  3 invoices

--- CRITICAL (81 days overdue) ---
To: Deputy Commissioner Accounts, Chennai Municipal Corporation
Re: FORMAL PAYMENT DEMAND - INV-B2B-1003

Dear Sir/Madam, Invoice INV-B2B-1003 for Rs 8,92,000 remains unpaid 81 days
after the due date of 15 January 2025. Despite previous communication, this
amount has not been settled. We formally request payment within 7 days.
Failure to respond may compel us to initiate legal recovery proceedings.
Transfer to ShopMax India - HDFC Bank IFSC HDFC0001234 Acc 50200012345678.

--- OVERDUE (37 days) ---
To: Ramesh Gupta, CFO, TechCorp Solutions
Re: Invoice INV-B2B-1001 - Payment Due

Dear Ramesh, Invoice INV-B2B-1001 for Rs 2,45,000 was due on 1 March 2025
under NET 30 terms and is now 37 days overdue. Please arrange payment at
your earliest convenience. UPI: shopmax@hdfcbank

--- REMINDER (23 days) ---
To: Sunita Rao, Accounts, Infosys BPO
Re: Friendly Reminder - INV-B2B-1002

Hi Sunita, a gentle reminder that INV-B2B-1002 for Rs 1,28,500 is now
23 days overdue. Please let us know if you need any documents.
UPI: shopmax@hdfcbank

Run this agent weekly via Cloud Scheduler. Log each chase attempt in Firestore with a timestamp so the agent tracks escalation history. Integrate with your email or WhatsApp Business API to dispatch messages automatically. Pause escalation for any invoice where a payment plan has been agreed and logged in Firestore.

Send your comments, suggestions or queries regarding this site to [email protected].