Creating an intelligent chatbot involves integrating advanced language models with effective development frameworks. In this session, we will explore how to build and run your first local chatbot instance using LangChain, a framework designed for developing applications powered by large language models (LLMs) in conjunction with IBM's Granite LLM model. LangChain streamlines the development process by offering open-source components and third-party integrations, enabling developers to create stateful agents featuring capabilities such as streaming and human-in-the-loop support. IBM's Granite LLM models offer various capabilities, including code generation, bug fixing, and code explanation across multiple programming languages. By combining these tools, participants will learn how to develop a responsive and intelligent chatbot that can understand and generate human-like text. The session will include practical steps, from setting up the development environment to running a functional local chatbot instance, providing participants with actionable insights to leverage these technologies in their own projects.

Scaling applications dynamically to meet varying demands can be challenging. OpenShift Serverless addresses this by providing Kubernetes-native building blocks, enabling developers to create and deploy serverless, event-driven applications. Leveraging the open-source Knative project, OpenShift Serverless offers features such as automatic scaling — including the ability to scale down to zero during periods of inactivity — efficient resource utilization, and seamless integration with event sources. This empowers developers to focus on writing code without managing the underlying infrastructure, leading to faster development cycles and reduced operational overhead. In this session, I will showcase how to build a lightweight yet highly scalable application using OpenShift Serverless's Serving and Eventing capabilities. Attendees will gain practical insights and actionable knowledge to leverage serverless technologies for their workloads.