AgentService

AgentService is a class that helps to create a collection of feature services from a list of tools, prompts, and knowledge bases.

Example

Let's build a chatbot helping a student to shop for items.


from tecton_gen_ai.testing import make_local_source, make_local_batch_feature_view, set_dev_mode
from tecton_gen_ai.testing.utils import make_local_vector_db_config

set_dev_mode()  # Set the dev mode to avoid tecton login

student = make_local_batch_feature_view(
    "student",
    {"student_id": 1, "name": "Jim", "teacher": "Mr. Smith", "preference": "fruit"},
    ["student_id"],
    description="Student information including name, teacher and shopping preference",
)

df = [
    {"zip":"98005", "item_id":1, "description":"pencil"},
    {"zip":"98005", "item_id":2, "description":"car"},
    {"zip":"98005", "item_id":3, "description":"paper"},
    {"zip":"10065", "item_id":4, "description":"boat"},
    {"zip":"10065", "item_id":5, "description":"cheese"},
    {"zip":"10065", "item_id":6, "description":"apple"},
]

src = make_local_source(
    "for_sale",
    df,
    description="Items information",  # required for source_as_knowledge
)
vdb_conf = make_local_vector_db_config()

# Create a knowledge base from the source
from tecton_gen_ai.fco import prompt, source_as_knowledge

@prompt(sources=[student])
def sys_prompt(student) -> str:
    return "You are serving a 4 years old child "+student["name"]

knowledge = source_as_knowledge(
    src,
    vector_db_config=vdb_conf,
    vectorize_column="description",
    filter = [("zip", str, "the zip code of the item for sale")]
)

# Serve the knowledge base
from tecton_gen_ai.fco import AgentService

service = AgentService(
    "app",
    prompts=[sys_prompt],
    tools=[student],
    knowledge=[knowledge],
)

# Test locally
from langchain_openai import ChatOpenAI

openai = ChatOpenAI(model = "gpt-4o", temperature=0)

from tecton_gen_ai.fco import AgentClient

client = AgentClient.from_local(service)

with client.set_context({"zip":"98005", "student_id":1}):
    print(client.invoke_agent(openai, "Suggest something for me to buy"))
with client.set_context({"zip":"10065", "student_id":1}):
    print(client.invoke_agent(openai, "Suggest something for me to buy"))

Methods

Name	Description
__init__(...)	Initialize AgentService

init(...)

Parameters

• name (str) - The name of the agent service


• prompts (Optional[List[Any]]) - A list of prompts Default: None


• tools (Optional[List[Any]]) - A list of tools Default: None


• knowledge (Optional[List[Any]]) - A list of knowledge bases Default: None


• feature_service_kwargs (Any) -

Example

Let's build a chatbot helping a student to shop for items.


from tecton_gen_ai.testing import make_local_source, make_local_batch_feature_view, set_dev_mode
from tecton_gen_ai.testing.utils import make_local_vector_db_config

set_dev_mode()  # Set the dev mode to avoid tecton login

student = make_local_batch_feature_view(
    "student",
    {"student_id": 1, "name": "Jim", "teacher": "Mr. Smith", "preference": "fruit"},
    ["student_id"],
    description="Student information including name, teacher and shopping preference",
)

df = [
    {"zip":"98005", "item_id":1, "description":"pencil"},
    {"zip":"98005", "item_id":2, "description":"car"},
    {"zip":"98005", "item_id":3, "description":"paper"},
    {"zip":"10065", "item_id":4, "description":"boat"},
    {"zip":"10065", "item_id":5, "description":"cheese"},
    {"zip":"10065", "item_id":6, "description":"apple"},
]

src = make_local_source(
    "for_sale",
    df,
    description="Items information",  # required for source_as_knowledge
)
vdb_conf = make_local_vector_db_config()

# Create a knowledge base from the source
from tecton_gen_ai.fco import prompt, source_as_knowledge

@prompt(sources=[student])
def sys_prompt(student) -> str:
    return "You are serving a 4 years old child "+student["name"]

knowledge = source_as_knowledge(
    src,
    vector_db_config=vdb_conf,
    vectorize_column="description",
    filter = [("zip", str, "the zip code of the item for sale")]
)

# Serve the knowledge base
from tecton_gen_ai.fco import AgentService

service = AgentService(
    "app",
    prompts=[sys_prompt],
    tools=[student],
    knowledge=[knowledge],
)

# Test locally
from langchain_openai import ChatOpenAI

openai = ChatOpenAI(model = "gpt-4o", temperature=0)

from tecton_gen_ai.fco import AgentClient

client = AgentClient.from_local(service)

with client.set_context({"zip":"98005", "student_id":1}):
    print(client.invoke_agent(openai, "Suggest something for me to buy"))
with client.set_context({"zip":"10065", "student_id":1}):
    print(client.invoke_agent(openai, "Suggest something for me to buy"))