refactor demos (#398)

demos/samples_python/human_resources_agent/Dockerfile

@@ -0,0 +1,16 @@
+FROM python:3.12 AS base
+
+FROM base AS builder
+
+WORKDIR /src
+
+COPY requirements.txt /src/
+RUN pip install --prefix=/runtime --force-reinstall -r requirements.txt
+
+FROM python:3.12-slim AS output
+COPY --from=builder /runtime /usr/local
+
+WORKDIR /app
+COPY . /app
+
+CMD ["uvicorn", "main:app", "--host", "0.0.0.0", "--port", "80", "--log-level", "info"]

demos/samples_python/human_resources_agent/README.md

@@ -0,0 +1,29 @@
+# HR Agent Demo
+
+This demo showcases how the **Arch** can be used to build an HR agent to manage workforce-related inquiries, workforce planning, and communication via Slack. It intelligently routes incoming prompts to the correct targets, providing concise and useful responses tailored for HR and workforce decision-making.
+
+## Available Functions:
+
+- **HR Q/A**: Handles general Q&A related to insurance policies.
+  - **Endpoint**: `/agent/hr_qa`
+
+- **Workforce Data Retrieval**: Retrieves data related to workforce metrics like headcount, satisfaction, and staffing.
+  - **Endpoint**: `/agent/workforce`
+  - Parameters:
+    - `staffing_type` (str, required): Type of staffing (e.g., `contract`, `fte`, `agency`).
+    - `region` (str, required): Region for which the data is requested (e.g., `asia`, `europe`, `americas`).
+    - `point_in_time` (int, optional): Time point for data retrieval (e.g., `0 days ago`, `30 days ago`).
+
+- **Initiate Policy**: Sends messages to a Slack channel
+  - **Endpoint**: `/agent/slack_message`
+  - Parameters:
+    - `slack_message` (str, required): The message content to be sent
+
+# Starting the demo
+1. Please make sure the [pre-requisites](https://github.com/katanemo/arch/?tab=readme-ov-file#prerequisites) are installed correctly
+2. Start Arch
+   ```sh
+   sh run_demo.sh
+   ```
+3. Navigate to http://localhost:18080/agent/chat
+4. "Can you give me workforce data for asia?"

demos/samples_python/human_resources_agent/arch_config.yaml

@@ -0,0 +1,62 @@
+version: v0.1
+listener:
+  address: 127.0.0.1
+  port: 8080 #If you configure port 443, you'll need to update the listener with tls_certificates
+  message_format: huggingface
+
+# Centralized way to manage LLMs, manage keys, retry logic, failover and limits in a central way
+llm_providers:
+  - name: OpenAI
+    provider_interface: openai
+    access_key: $OPENAI_API_KEY
+    model: gpt-4o-mini
+    default: true
+
+# Arch creates a round-robin load balancing between different endpoints, managed via the cluster subsystem.
+endpoints:
+  app_server:
+    # value could be ip address or a hostname with port
+    # this could also be a list of endpoints for load balancing
+    # for example endpoint: [ ip1:port, ip2:port ]
+    endpoint: host.docker.internal:18080
+    # max time to wait for a connection to be established
+    connect_timeout: 0.005s
+
+# default system prompt used by all prompt targets
+system_prompt: |
+  You are a Workforce assistant that helps on workforce planning and HR decision makers with reporting and workforce planning. Use following rules when responding,
+  - when you get data in json format, offer some summary but don't be too verbose
+  - be concise, to the point and do not over analyze the data
+
+prompt_targets:
+    - name: workforce
+      description: Get workforce data like headcount and satisfaction levels by region and staffing type
+      endpoint:
+        name: app_server
+        path: /agent/workforce
+        http_method: POST
+      parameters:
+        - name: staffing_type
+          type: str
+          description: specific category or nature of employment used by an organization like fte, contract and agency
+          required: true
+          enum: [fte, contract, agency]
+        - name: region
+          type: str
+          required: true
+          description: Geographical region for which you want workforce data like asia, europe, americas.
+        - name: data_snapshot_days_ago
+          type: int
+          required: false
+          description: the snapshot day for which you want workforce data.
+    - name: slack_message
+      endpoint:
+        name: app_server
+        path: /agent/slack_message
+        http_method: POST
+      description: sends a slack message on a channel
+      parameters:
+        - name: slack_message
+          type: string
+          required: true
+          description: the message that should be sent to a slack channel

demos/samples_python/human_resources_agent/docker-compose.yaml

@@ -0,0 +1,29 @@
+services:
+  api_server:
+    build:
+      context: .
+    environment:
+      - SLACK_BOT_TOKEN=${SLACK_BOT_TOKEN:-None}
+      - OPENAI_API_KEY=${OPENAI_API_KEY:?error}
+      - CHAT_COMPLETION_ENDPOINT=http://host.docker.internal:10000/v1
+    volumes:
+      - ./arch_config.yaml:/app/arch_config.yaml
+    ports:
+      - "18080:80"
+    healthcheck:
+        test: ["CMD", "curl" ,"http://localhost:80/healthz"]
+        interval: 5s
+        retries: 20
+
+  chatbot_ui:
+    build:
+      context: ../../shared/chatbot_ui
+      dockerfile: Dockerfile
+    ports:
+      - "18080:8080"
+    environment:
+      - CHAT_COMPLETION_ENDPOINT=http://host.docker.internal:10000/v1
+    extra_hosts:
+      - "host.docker.internal:host-gateway"
+    volumes:
+      - ./arch_config.yaml:/app/arch_config.yaml

demos/samples_python/human_resources_agent/main.py

@@ -0,0 +1,94 @@
+import os
+import json
+import pandas as pd
+import gradio as gr
+import logging
+
+from fastapi import FastAPI, HTTPException
+from pydantic import BaseModel, Field
+from enum import Enum
+from typing import List, Optional, Tuple
+from slack_sdk import WebClient
+from slack_sdk.errors import SlackApiError
+from openai import OpenAI
+
+app = FastAPI()
+workforce_data_df = None
+
+with open("workforce_data.json") as file:
+    workforce_data = json.load(file)
+    workforce_data_df = pd.json_normalize(
+        workforce_data,
+        record_path=["regions"],
+        meta=["data_snapshot_days_ago", "satisfaction"],
+    )
+
+
+# Define the request model
+class WorkforceRequest(BaseModel):
+    region: str
+    staffing_type: str
+    data_snapshot_days_ago: Optional[int] = None
+
+
+class SlackRequest(BaseModel):
+    slack_message: str
+
+
+class WorkforceResponse(BaseModel):
+    region: str
+    staffing_type: str
+    headcount: int
+    satisfaction: float
+
+
+@app.post("/agent/slack_message")
+def send_slack_message(request: SlackRequest):
+    """
+    Endpoint that sends slack message
+    """
+    slack_message = request.slack_message
+
+    # Load the bot token from an environment variable or replace it directly
+    slack_token = os.getenv(
+        "SLACK_BOT_TOKEN"
+    )  # Replace with your token if needed: 'xoxb-your-token'
+
+    if slack_token is None:
+        print(f"Message for slack: {slack_message}")
+    else:
+        client = WebClient(token=slack_token)
+        channel = "hr_agent_demo"
+        try:
+            # Send the message
+            response = client.chat_postMessage(channel=channel, text=slack_message)
+            return f"Message sent to {channel}: {response['message']['text']}"
+        except SlackApiError as e:
+            print(f"Error sending message: {e.response['error']}")
+
+
+# Post method for device summary
+@app.post("/agent/workforce")
+def get_workforce(request: WorkforceRequest):
+    """
+    Endpoint to workforce data by region, staffing type at a given point in time.
+    """
+    region = request.region.lower()
+    staffing_type = request.staffing_type.lower()
+    data_snapshot_days_ago = (
+        request.data_snapshot_days_ago
+        if request.data_snapshot_days_ago
+        else 0  # this param is not required.
+    )
+
+    response = {
+        "region": region,
+        "staffing_type": f"Staffing agency: {staffing_type}",
+        "headcount": f"Headcount: {int(workforce_data_df[(workforce_data_df['region']==region) & (workforce_data_df['data_snapshot_days_ago']==data_snapshot_days_ago)][staffing_type].values[0])}",
+        "satisfaction": f"Satisfaction: {float(workforce_data_df[(workforce_data_df['region']==region) & (workforce_data_df['data_snapshot_days_ago']==data_snapshot_days_ago)]['satisfaction'].values[0])}",
+    }
+    return response
+
+
+if __name__ == "__main__":
+    app.run(debug=True)

demos/samples_python/human_resources_agent/requirements.txt

@@ -0,0 +1,13 @@
+fastapi
+uvicorn
+slack-sdk
+typing
+pandas
+gradio==5.3.0
+async_timeout==4.0.3
+loguru==0.7.2
+asyncio==3.4.3
+httpx==0.27.0
+python-dotenv==1.0.1
+pydantic==2.8.2
+openai==1.51.0

demos/samples_python/human_resources_agent/run_demo.sh

@@ -0,0 +1,47 @@
+#!/bin/bash
+set -e
+
+# Function to start the demo
+start_demo() {
+  # Step 1: Check if .env file exists
+  if [ -f ".env" ]; then
+    echo ".env file already exists. Skipping creation."
+  else
+    # Step 2: Create `.env` file and set OpenAI key
+    if [ -z "$OPENAI_API_KEY" ]; then
+      echo "Error: OPENAI_API_KEY environment variable is not set for the demo."
+      exit 1
+    fi
+
+    echo "Creating .env file..."
+    echo "OPENAI_API_KEY=$OPENAI_API_KEY" > .env
+    echo ".env file created with OPENAI_API_KEY."
+  fi
+
+  # Step 3: Start Arch
+  echo "Starting Arch with arch_config.yaml..."
+  archgw up arch_config.yaml
+
+  # Step 4: Start Network Agent
+  echo "Starting HR Agent using Docker Compose..."
+  docker compose up -d  # Run in detached mode
+}
+
+# Function to stop the demo
+stop_demo() {
+  # Step 1: Stop Docker Compose services
+  echo "Stopping HR Agent using Docker Compose..."
+  docker compose down
+
+  # Step 2: Stop Arch
+  echo "Stopping Arch..."
+  archgw down
+}
+
+# Main script logic
+if [ "$1" == "down" ]; then
+  stop_demo
+else
+  # Default action is to bring the demo up
+  start_demo
+fi

demos/samples_python/human_resources_agent/test_data.yaml

@@ -0,0 +1,14 @@
+test_cases:
+  - id: get workforce data
+    input:
+      messages:
+        - role: user
+          content: what is workforce data for asia for fte employees
+    expected_tools:
+      - type: function
+        function:
+          name: workforce
+          arguments:
+            staffing_type: fte
+            region: asia
+    expected_output_contains: asia

demos/samples_python/human_resources_agent/workforce_data.json

@@ -0,0 +1,29 @@
+[
+    {
+        "data_snapshot_days_ago": 0,
+        "regions": [
+            { "region": "asia", "contract": 100, "fte": 150, "agency": 2000 },
+            { "region": "europe", "contract": 80, "fte": 120, "agency": 2500 },
+            { "region": "americas", "contract": 90, "fte": 200, "agency": 3100 }
+        ],
+        "satisfaction": 3.5
+    },
+    {
+        "data_snapshot_days_ago": 30,
+        "regions": [
+            { "region": "asia", "contract": 110, "fte": 155, "agency": 1000 },
+            { "region": "europe", "contract": 85, "fte": 130, "agency": 1600 },
+            { "region": "americas", "contract": 95, "fte": 210, "agency": 3100 }
+        ],
+        "satisfaction": 4.0
+    },
+    {
+        "data_snapshot_days_ago": 60,
+        "regions": [
+            { "region": "asia", "contract": 115, "fte": 160, "agency": 500 },
+            { "region": "europe", "contract": 90, "fte": 140, "agency": 700 },
+            { "region": "americas", "contract": 100, "fte": 220, "agency": 1200 }
+        ],
+        "satisfaction": 4.7
+    }
+]