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

	Facebook
	Twitter
	Linkedin
	E-Mail

Generative AI > Google Gemini API > Product Configurator Agent

Product Configurator Agent

Author: Venkata Sudhakar

ShopMax India sells complex IT infrastructure solutions where selecting the wrong components leads to incompatible configurations and costly returns. A Product Configurator Agent acts as a guided selling tool - it asks the right questions about the customer needs, recommends compatible components, flags conflicts, and outputs a complete bill of materials ready for quoting.

This tutorial builds a Product Configurator Agent using Google ADK and Gemini 2.0 Flash. The agent takes customer workload requirements, selects compatible server, storage, and networking components, validates the configuration, and produces a BOM with pricing.

The below example shows the Product Configurator Agent building a server configuration for a mid-size manufacturing company in Coimbatore.

from google.adk.agents import Agent
from google.adk.tools import FunctionTool
from google.adk.runners import Runner
from google.adk.sessions import InMemorySessionService
from google.genai.types import Content, Part

# Component catalogue
SERVERS = {
    "small":  {"sku": "SVR-S1", "name": "Dell PowerEdge T150", "cores": 4,  "ram_gb": 16,  "price_lakh": 1.8},
    "medium": {"sku": "SVR-M1", "name": "Dell PowerEdge R350", "cores": 8,  "ram_gb": 32,  "price_lakh": 3.2},
    "large":  {"sku": "SVR-L1", "name": "Dell PowerEdge R650", "cores": 16, "ram_gb": 64,  "price_lakh": 6.5},
}
STORAGE = {
    "1tb":  {"sku": "STG-1T", "name": "1TB SSD Array",  "price_lakh": 0.45},
    "4tb":  {"sku": "STG-4T", "name": "4TB SSD Array",  "price_lakh": 1.20},
    "12tb": {"sku": "STG-12T", "name": "12TB NAS Array", "price_lakh": 2.80},
}
NETWORKING = {
    "basic":    {"sku": "NET-B1", "name": "24-port 1GbE Switch",  "price_lakh": 0.35},
    "advanced": {"sku": "NET-A1", "name": "24-port 10GbE Switch", "price_lakh": 0.95},
}

def recommend_configuration(users: int, workload_type: str,
                             data_tb: float, ha_required: bool) -> dict:
    """Recommend server, storage, and networking based on requirements."""
    # Server selection
    if users <= 20 and workload_type == "office":
        server_size = "small"
    elif users <= 100 or workload_type in ["erp", "database"]:
        server_size = "medium"
    else:
        server_size = "large"
    server = SERVERS[server_size]

# Storage selection
    if data_tb <= 1:
        storage_key = "1tb"
    elif data_tb <= 4:
        storage_key = "4tb"
    else:
        storage_key = "12tb"
    storage = STORAGE[storage_key]
    storage_qty = 2 if ha_required else 1  # RAID pair for HA

# Networking
    net_key = "advanced" if workload_type in ["database", "erp"] else "basic"
    network = NETWORKING[net_key]

# UPS - always recommended
    ups = {"sku": "UPS-2K", "name": "APC 2KVA UPS", "price_lakh": 0.18, "qty": 1}

total = (server["price_lakh"] +
             storage["price_lakh"] * storage_qty +
             network["price_lakh"] +
             ups["price_lakh"])

return {
        "recommended_config": {
            "server": {**server, "qty": 1},
            "storage": {**storage, "qty": storage_qty,
                        "note": "RAID-1 pair for high availability" if ha_required else ""},
            "networking": {**network, "qty": 1},
            "ups": ups
        },
        "total_price_lakh": round(total, 2),
        "ha_configured": ha_required,
        "workload": workload_type,
        "users_supported": users
    }

def validate_compatibility(server_sku: str, storage_sku: str) -> dict:
    """Check if server and storage combination is compatible."""
    # SVR-S1 (tower) does not support NAS arrays
    incompatible = (server_sku == "SVR-S1" and storage_sku == "STG-12T")
    return {
        "compatible": not incompatible,
        "issue": "Tower server SVR-S1 does not support 12TB NAS array. Upgrade to rack server." if incompatible else None
    }

config_tool = FunctionTool(func=recommend_configuration)
validate_tool = FunctionTool(func=validate_compatibility)

configurator_agent = Agent(
    model="gemini-2.0-flash",
    name="product_configurator_agent",
    instruction="""
        You are a Product Configurator Agent for ShopMax India.
        When given customer requirements:
        1. Recommend an appropriate server configuration
        2. Validate component compatibility
        3. Output a complete Bill of Materials (BOM):
           - Each component with SKU, name, qty, unit price, line total
           - Configuration notes (HA setup, workload fit)
           - Total solution price in Rs lakhs
           - Any compatibility warnings
           - Optional add-ons to consider
        Present the BOM clearly. Flag any issues before the customer commits.
    """,
    tools=[config_tool, validate_tool]
)

session_service = InMemorySessionService()
runner = Runner(agent=configurator_agent, app_name="configurator", session_service=session_service)

session = session_service.create_session(app_name="configurator", user_id="sales_rep")
prompt = """
    Configure a server solution for:
    Customer: Coimbatore Textile Mills
    Users: 60
    Workload: ERP system (SAP B1)
    Data storage needed: 3TB
    High availability required: Yes
"""
message = Content(role="user", parts=[Part(text=prompt)])
for event in runner.run(user_id="sales_rep", session_id=session.id, new_message=message):
    if event.is_final_response():
        print(event.content.parts[0].text)

It gives the following output,

BILL OF MATERIALS - Coimbatore Textile Mills
=============================================
Workload : ERP (SAP B1) | Users: 60 | HA: Yes

Component  SKU       Description              Qty  Unit Price   Total
Server     SVR-M1    Dell PowerEdge R350       1   Rs 3.20 L    Rs 3.20 L
Storage    STG-4T    4TB SSD Array             2   Rs 1.20 L    Rs 2.40 L
                     (RAID-1 pair for HA)
Networking NET-A1    24-port 10GbE Switch      1   Rs 0.95 L    Rs 0.95 L
UPS        UPS-2K    APC 2KVA UPS              1   Rs 0.18 L    Rs 0.18 L

Total Solution Price: Rs 6.73 lakhs (excl. GST)

Configuration Notes
  - R350 with 8 cores and 32GB RAM is well-suited for SAP B1
    with 60 concurrent users
  - Dual 4TB SSD arrays in RAID-1 provide HA with 4TB usable storage
  - 10GbE switching recommended for ERP database performance

Compatibility: All components validated - no conflicts

Recommended Add-ons
  - 3-year ProSupport Plus (Dell) - Rs 0.85 L
  - Offsite backup solution     - Rs 0.40 L/yr

The Product Configurator Agent eliminates misconfigured orders and reduces pre-sales engineering time. Extend it to support GPU add-ons for AI workloads, rack space planning, and cooling requirements. Integrate with your ERP to check real-time component availability before committing to the BOM.

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