AI powered expert system demo

1. $ git clone this repo && $ cd into root of it.
2. Create all defined volumes $ apolo-flow mkvolumes
3. Upload docs $ apolo-flow upload docs if we don't use ${{ upload(volumes.docs).ref_ro }} in the live.yml
4. Build image for web app with $ apolo-flow build pfb
5. $ apolo-flow run pgvector -- start postgresql.
6. $ apolo-flow run ollama -- start Ollama with VLLM server under the hood
7. $ apolo-flow run pfb -- start application.

Query examples

Get booking details - booking number: 102, first: Jane, last: Smith
ChangeBooking -  102, Jane, Smith, 2025-11-25, from: FUN, to: FUN
CancelBooking -  booking number: 102, first: Jane, last: Smith

Run ollama locally:

1. Run ollama in docker https://hub.docker.com/r/ollama/ollama
2. Run qwen2:1.5b model qwen2 with 1.5b parameters locally - docker exec -it ollama ollama run qwen2:1.5b

Forward ollama:

apolo job port-forward {job name} {port}:{port}
apolo job port-forward playground-flight-booking-ollama 11434:11434

Forward DB:

apolo job port-forward playground-flight-booking-pgvector 5432:5432
psql -h 0.0.0.0 -U postgres -d postgres

Rest http call:

POST http://localhost:11434/api/generate
Accept: application/json

{
"model": "llama3.2:1b",
"prompt": "What color is the sky at different times of the day? Respond using JSON",
"format": "json",
"stream": false
}

Spring AI re-implementation of https://github.com/marcushellberg/java-ai-playground

This app shows how you can use Spring AI to build an AI-powered system that:

• Has access to terms and conditions (retrieval augmented generation, RAG)
• Can access tools (Java methods) to perform actions (Function Calling)
• Uses an LLM to interact with the user

Requirements

• Java 17+
• OpenAI API key in OPENAI_API_KEY environment variable

Running

Run the app by running Application.java in your IDE or mvn in the command line.

With OpenAI Chat

Add to the POM the Spring AI Open AI boot starter:

<dependency>
    <groupId>org.springframework.ai</groupId>
    <artifactId>spring-ai-openai-spring-boot-starter</artifactId>
</dependency>

Add the OpenAI configuraiton to the applicaiton.properties:

spring.ai.openai.api-key=${OPENAI_API_KEY}
spring.ai.openai.chat.options.model=gpt-4o

WIth VertexAI Geminie Chat

Add to the POM the Spring AI VertexAI Gemeni and Onnx Transfomer Embedding boot starters:

<dependency>
    <groupId>org.springframework.ai</groupId>
    <artifactId>spring-ai-vertex-ai-gemini-spring-boot-starter</artifactId>
</dependency>

<dependency>
    <groupId>org.springframework.ai</groupId>
    <artifactId>spring-ai-transformers-spring-boot-starter</artifactId>
</dependency>

Add the VertexAI Gemini configuraiton to the applicaiton.properties:

spring.ai.vertex.ai.gemini.project-id=${VERTEX_AI_GEMINI_PROJECT_ID}
spring.ai.vertex.ai.gemini.location=${VERTEX_AI_GEMINI_LOCATION}
spring.ai.vertex.ai.gemini.chat.options.model=gemini-1.5-pro-001
# spring.ai.vertex.ai.gemini.chat.options.model=gemini-1.5-flash-001

With Azure OpenAI Chat

Add to the POM the Spring AI Azure OpenAI boot starter:

<dependency>
    <groupId>org.springframework.ai</groupId>
    <artifactId>spring-ai-azure-openai-spring-boot-starter</artifactId>
</dependency>

Add the Azure OpenAI configuraiton to the applicaiton.properties:

spring.ai.azure.openai.api-key=${AZURE_OPENAI_API_KEY}
spring.ai.azure.openai.endpoint=${AZURE_OPENAI_ENDPOINT}
spring.ai.azure.openai.chat.options.deployment-name=gpt-4o

With Groq Chat

It reuses the OpenAI Chat client but ponted to the Groq endpont

Add to the POM the Spring AI Open AI boot starter:

<dependency>
    <groupId>org.springframework.ai</groupId>
    <artifactId>spring-ai-openai-spring-boot-starter</artifactId>
</dependency>
<dependency>
    <groupId>org.springframework.ai</groupId>
    <artifactId>spring-ai-transformers-spring-boot-starter</artifactId>
</dependency>

Add the Groq configuraiton to the applicaiton.properties:

spring.ai.openai.api-key=${GROQ_API_KEY}
spring.ai.openai.base-url=https://api.groq.com/openai
spring.ai.openai.chat.options.model=llama3-70b-8192

With Anthropic Claude 3 Chat

Add to the POM the Spring AI Anthropic Claude and Onnx Transfomer Embedding boot starters:

<dependency>
    <groupId>org.springframework.ai</groupId>
    <artifactId>spring-ai-anthropic-spring-boot-starter</artifactId>
</dependency>

<dependency>
    <groupId>org.springframework.ai</groupId>
    <artifactId>spring-ai-transformers-spring-boot-starter</artifactId>
</dependency>

Add the Anthropic configuraiton to the applicaiton.properties:

spring.ai.anthropic.api-key=${ANTHROPIC_API_KEY}
spring.ai.openai.chat.options.model=llama3-70b-8192
spring.ai.anthropic.chat.options.model=claude-3-5-sonnet-20240620

Build Jar

./mvnw clean install -Pproduction

java -jar ./target/playground-flight-booking-0.0.1-SNAPSHOT.jar

docker run -it --rm --name postgres -p 5432:5432 -e POSTGRES_USER=postgres -e POSTGRES_PASSWORD=postgres ankane/pgvector