I want to share my experience in the cloud, AI, and agents development areas.
The post discusses integrating Knowledge Graphs with Retrieval-Augmented Generation (RAG), specifically using Neo4j and LangGraph. It outlines an example setup where extracted document data forms a structured graph for querying. The system enables natural question-and-answer interactions through AI, enhancing information retrieval with graph relationships and embeddings.
The blog post outlines the process of setting up the Agent Developer Kit (ADK) to build and run AI agents locally using WatsonX Orchestrate Developer Edition. It involves setting up prerequisites, installing the necessary software, and loading an example agent—optional integration with Langfuse for observability.
The post explores the integration of Large Language Models (LLMs) in applications, stressing the need for effective production testing. It introduces the AI Operational Complexity Cube concept, emphasizing new testing dimensions for LLMs, including prompt testing and user engagement. A structured testing approach is proposed to ensure reliability and robustness.
This blog post details the development and deployment of a customizable AI Agent using watsonx.ai. It covers motivations, architecture, and code for a weather query tool, explaining local execution, testing with pytest, and deployment via scripts. The integration with Streamlit UI is emphasized, showcasing seamless deployment processes and enhanced functionality for developers.
This blog post explains how to deploy a fine-tuned model to IBM watsonx on IBM Cloud. It highlights the advantages of using this platform, such as avoiding infrastructure management and ensuring enterprise security, as well as detailed steps for configuration, deployment, and accessing the model from IBM watsonx.
This blog post provides updates on the InstructLab project by IBM and Red Hat, detailing installation and configuration changes. It discusses new default locations for files and troubleshooting steps for model serving, emphasizing an overall installation process that remains largely consistent with prior guidance while noting minor user-friendly adjustments.
The blog post outlines updates on fine-tuning a model with the InstructLab , detailing tasks like data preparation, validation, synthetic data generation, model training, and testing. It emphasizes the need for extensive and accurate input for effective training, while only minimal changes in the overall process since previous versions, particularly in handling data quality. This blog post contains updates related to my blog post InstructLab and Taxonomy tree: LLM Foundation Model Fine-tuning Guide | Musician Example.
This blog post discusses the creation of a custom Bee Agent that operates independently from the Bee Stack and interacts in German. It explores requirements, agent examples, coding in TypeScript, and GitHub references. The author implements an agent using a specific system prompt while addressing the challenges of ensuring consistent output in German.
This blog post describes a web chat application using a large language model on watsonx, with the interface built in Streamlit.io. It focuses on motivation, architecture, code sections, and local setup, featuring basic authentication and options for user interaction. The author highlights Streamlit’s rapid prototyping capabilities and ease of use with Python.
In 2024, I published 35 blog posts focusing on innovation in AI and development, emphasizing the Bee Agent Framework and IBM's watsonx.ai. The highlights including creating cheat sheets, exploring large language models, and providing practical guides. That boost my excitement for future challenges in 2025.
Thomas Suedbroecker's Blog 