Skip to content

Requirements

Software Requirements

AI Model Requirements

Using OpenAI

All of the examples in this workshop use OpenAI by default to serve the LLM that is used to build our application. If you want to use them “as-is”, you will need an OpenAI API key to complete this workshop.

If you do not already have one, create an API key.

No instructor-provided key?

New OpenAI developer accounts receive $5 in free trial credits.
If you already used your credits, you’ll need to fund your account.

Tip

Don’t worry — this workshop is inexpensive. The total cost should not exceed $0.50 (~€0.43).
See the OpenAI pricing calculator.

Once you have a key, set it as an environment variable:

export OPENAI_API_KEY=<your-key>

$Env:OPENAI_API_KEY = <your-key>

Using other models

If you do not want to use OpenAI to serve the LLM, LangChain4j and LangChain4j CDI makes it straightforward to integrate any other service providers. For instance we could serve our model on our local machine using an Ollama server.

The applications that you will build are configured using the src/main/resources/META-INF/microprofile-config.properties file. Each one will include an example base-url property that can be used to configure the agent to connect to a specific LLM:

application.properties
# If you want to use a different provider or run an LLM on your local machine,
# uncomment this line and update the url/port accordingly.
# dev.langchain4j.cdi.plugin.customer-support-agent.config.base-url=http://localhost:11434/v1

Simply uncomment this line and modify the value of the base-url property to point at your own LLM. You may also need to specify an API key for your model and the model that is being served. In order to do this, modify the api-key and model-name properties defined in the src/main/resources/META-INF/microprofile-config.properties file. For example:

application.properties
dev.langchain4j.cdi.plugin.customer-support-agent.config.api-key=${MY_API_KEY}
dev.langchain4j.cdi.plugin.customer-support-agent.config.model-name=gpt-oss:20b

Good to Know

Liberty Dev Mode

All of the examples in this workshop use Open Liberty to run the agent applications. You can run the applications in dev mode from the project directory:

./mvnw liberty:dev

Dev mode automatically recompiles your code on every change. Your app will be available at http://localhost:9080/.

Switching steps

Stop the running application (Ctrl+C) before starting the next step.

Debugging

To debug an app in dev mode, put breakpoints in your code and attach your IDE debugger. In VSCode, use the Liberty Tools extension. With the application running in Open Liberty, right click on the application in the LIBERTY DASHBOARD view in the explorer and select Attach debugger. Other IDEs (Eclipse, IntelliJ) support similar remote debugging.

Getting the Workshop Material

Either clone the repository with Git or download a ZIP archive.

With Git

git clone https://github.com/msmiths/langchain4j-workshop.git
cd langchain4j-workshop

Direct Download

curl -L -o workshop.zip https://github.com/msmiths/langchain4j-workshop/archive/refs/heads/main.zip
unzip workshop.zip
cd langchain4j-workshop-main

Pre-Warming Caches

This workshop requires downloading Maven dependencies and Docker images. To avoid bandwidth issues during the session, we recommend pre-downloading them.

Warm up Maven

./mvnw verify

Tip

This command not only downloads dependencies but also verifies your setup before the workshop.

Warm up Docker Images

  • Podman:
    • podman pull pgvector/pgvector:pg17
    • podman pull grafana/otel-lgtm
  • Docker:
    • docker pull pgvector/pgvector:pg17
    • docker pull grafana/otel-lgtm

Importing the Project in Your IDE

Tip

Open the project from section-1/step-01 in your IDE and use that directory throughout the workshop.

If you get stuck, simply switch to the step-xx directory of the last completed step.

Next Step

Once ready, you can pick one of these entries points to start the workshop: