Kotlin AI Examples is a repository showcasing various AI frameworks integrated into Kotlin-based projects.
Here you’ll find ready-to-use examples for Spring AI, LangChain4j, as well as interactive Kotlin notebooks.
 Kotlin Projects |
 Kotlin Notebooks |
🚀 Getting Started | ⚙ Prerequisites |
Kotlin Projects (/projects)
This section contains complete Kotlin projects demonstrating AI integrations.
Spring AI Examples (/projects/spring-ai)
- helloworld: Basic example of using Spring AI
- spring-ai-examples: A comprehensive set of Spring AI feature demonstrations and examples
- playground-flight-booking: An AI-powered flight booking system demo using multiple providers (OpenAI, VertexAI Gemini, Azure OpenAI, Groq, Anthropic Claude)
- spring-ai-mcp-server-example: spring mcp server sample
- springAI-demo: a Spring Boot application with Spring AI and Kotlin that loads Kotlin standard library documents into a Qdrant vector store, implements endpoints for similarity search and a RAG-powered chat interface, and integrates an LLM for detailed, document-driven answers.
LangChain4j Examples (/projects/langchain4j)
- langchain4j-spring-boot: Examples of using LangChain4j with Kotlin and Spring
MCP Examples (/projects/mcp)
- mcp-demo: this tutorial briefly describes the process of creating an MCP server in Kotlin to work with stock data from FMP, connecting it to Claude for Desktop, and developing a custom client based on OpenAI using Compose.
Kotlin Notebooks (/notebooks)
A collection of interactive Jupyter notebooks in Kotlin, organized by project.
Spring AI Kotlin Notebooks (/notebooks/spring-ai)
-
SpringAI_Overview.ipynb: overview of Spring AI with Kotlin
-
Tutorial Series This series of interactive notebooks provides a hands-on journey through Spring AI capabilities with Kotlin. Each notebook builds on the previous one, guiding you from basic concepts to advanced AI integrations.
Notebook Description 1. Intro Introduction to Spring AI fundamentals and setup with Kotlin 2. Prompts Working with prompt engineering techniques and best practices 3. Streaming Implementing streaming responses for real-time AI interactions 4. Tools Connecting LLMs to external tools and functions for enhanced capabilities 5. Structured Outputs Getting formatted, structured data from LLMs for easier application integration 6. Advisors Using Spring AI's Advisor system to enhance and customize AI interactions 7. RAG Implementing Retrieval-Augmented Generation with your own data sources 8. Text-to-Image Converting text descriptions into generated images with Spring AI 9. Text-to-Audio Creating speech from text and transcribing audio with AI models 10. Local Model Running AI models locally on your machine using Ollama and Spring AI
AI Agents Kotlin Notebooks (/notebooks/agents)
| Notebook | Description |
|---|---|
| Prompt Chaining | Implements a sequential workflow that breaks complex tasks into simpler steps, where each LLM call processes the output of the previous one |
| Routing | Shows how to classify inputs and direct them to specialized handlers, optimizing for different types of requests |
| Parallelization | Demonstrates running multiple LLM tasks simultaneously using Kotlin coroutines for improved performance |
| Orchestrator-Workers | Implements a pattern where a central LLM breaks down tasks dynamically and delegates them to specialized worker LLMs |
| Evaluator-Optimizer | Creates a feedback loop where one LLM generates responses and another evaluates them for continuous improvement |
LangChain4j Kotlin Notebooks (/notebooks/langchain4j)
- LangChain4j_Overview.ipynb: overview of LangChain4j with Kotlin, demonstrating how to work with chat models, manage prompts, streaming, and produce structured responses.
- SummarizingDocuments.ipynb: demonstrates how to split large text documents into smaller chunks, summarize each chunk with an AI model, and then merge the summaries into a concise final result.
Agents ReaCtor (ARC) Kotlin Notebooks (/notebooks/arc)
- WeatherAgent.ipynb: demonstrates creating an agent that retrieves real-time weather data for a specified location via WeatherAPI
- SummarizerAgent.ipynb: demonstrates creating an agent that summarizes web pages (e.g., blog posts) by processing HTML and generating concise summaries
KInference Kotlin Notebooks (/notebooks/kInference)
- KIClassification.ipynb: how to set up a classification environment using the KIEngine framework, manage cached data files, and perform inference on input data.
- KIGPT2.ipynb: how to configure and run a GPT-2 model with KIEngine, handling tokenization, caching, and text generation.
- ORTClassification.ipynb: how to perform classification using ORT-based inference, handling data loading, caching, and model execution steps.
- ORTGPT2.ipynb: how to run GPT-2 inference using the ORT engine, handling tokenization, caching, and generation of text outputs.
xef.ai Kotlin Notebooks (/notebooks/xefAI)
- xefAI_Overview.ipynb: xef.ai overview notebook
Each project in the projects directory has its own README with detailed instructions on how to run and use
the examples.
- Java 17+
- Kotlin
- Appropriate AI provider API keys (see each project’s README for more details)
Note
Make sure you have the necessary access credentials for your chosen AI service (OpenAI, Azure, VertexAI, etc.) and the required dependencies in your build scripts (Gradle/Maven).
⭐ If you enjoy this repository, please give it a star!
This project is licensed under the Apache License 2.0.