← Back to all guides

Quickstart — Build Your First AI Chatbot in 5 Minutes

Langoedge Team4 min read

Build Your First AI Chatbot

Ready to automate your first workflow? Follow these steps to build a Text Agent with Langoedge and deploy it to your website.

Key Takeaway: You can go from zero to a production-ready AI chatbot in under 5 minutes using Langoedge's visual graph builder.

1

Create a New Graph

Navigate to the **Graphs** tab in your Langoedge dashboard and click **New Graph**. Give your graph a descriptive name like "Customer Support Bot" or "Sales Qualification Agent".
2

Define the Personality

In the first Node (the Greeting Node), set your **System Prompt**. This is the personality and instructions for your AI agent.
*Example: "You are a helpful assistant for Acme Corp. Use our documentation to answer support queries. Always be polite and concise."*
3

Add a Knowledge Base (Optional)

If you want your agent to know about your company products, upload a PDF to the **Knowledge Base** and add a **Retriever Tool** (RAG) to your graph. This lets the AI search your documents before answering.
4

Add Tool Integrations (Optional)

Connect your agent to external tools — CRM webhooks, email services, or search APIs — by adding Tool Nodes and authenticating via the **Connect** tab.
5

Test and Deploy

Click **Embed / Deploy** and copy the JavaScript snippet. Paste it into your website's HTML to see the AI chat widget appear instantly.
> **Finding your Graph ID:** Once you're ready to invoke this agent from an external app, you'll need the `graph_config_id`. You can find it in the browser URL when the graph is open (e.g. `app.langoedge.com/graphs/graph_cfg_987654`) or in the graph's **Settings** panel. See the [External Integration guide](/guides/consume-external) for full API usage.

Example: The "Human Handoff"

You can build a logic branch that says: "If the user is frustrated, escalate to a human."

graph LR A[User Input] --> B{Check Sentiment} B -- Angry --> C[Slack Notification Node] B -- Happy --> D[Continue Helpful Chat]

This uses a Conditional Edge — a routing mechanism where the AI evaluates sentiment and directs the conversation to different paths based on the result.


Going from Text to Voice

Once you have a Text Graph, you can easily turn it into a Voice Agent that handles phone calls.


Common Quickstart Scenarios

Scenario 1: FAQ Chatbot

  • Goal: Answer customer questions from your knowledge base.
  • Setup: One LLM node + one Retriever Tool (RAG) + Knowledge Base PDF upload.

Scenario 2: Lead Qualification Bot

  • Goal: Score inbound website visitors and route hot leads to your sales team.
  • Setup: LLM node for scoring + Slack notification tool + CRM webhook.

Scenario 3: Support Triage Agent

  • Goal: Categorize support tickets by urgency and department.
  • Setup: Evaluation node + conditional edges for routing + email send tool.


Frequently Asked Questions

What if my chatbot doesn't respond correctly?
Open your graph in the Langoedge dashboard and use **Debug Mode** to step through each node execution. You can inspect what the LLM received and returned at every step, then adjust your system prompt or add conditional edges to handle edge cases.
Can I edit my graph after deploying it?
Yes. Changes to your graph are automatically compiled and deployed. There is no separate redeploy step — save your changes and the updated agent is live immediately.
How do I add more tools to my chatbot later?
Navigate to your graph, click **Add Node**, and select a Tool Node type. Authenticate the integration in the **Connect** tab, wire the node into your flow, and save. Your agent will start using the new tool on the next conversation.
Can I turn this text chatbot into a voice agent?
Yes. Create a Voice Graph in the dashboard, bind a phone number in the settings panel, and attach your existing Text Graph as a tool. Your voice agent can then call your text logic for complex tasks during phone conversations.

LT

Langoedge Team

The Langoedge engineering team builds AI agent infrastructure that empowers businesses to deploy reliable, observable AI staff. Follow Langoedge Team on LinkedIn for product updates and architectural deep dives.