Phillip from Base10 discusses the launch of "Chains," a new feature that allows users to link multiple AI models together and orchestrate workflows efficiently. This product enables developers to separate business logic from model code, making it easier to create complex AI solutions. One practical example is using "Chains" to transcribe large audio files in parallel by breaking them into smaller chunks, then stitching the results together—solving the challenge of processing long recordings. This multi-step, multimodel approach also supports advanced use cases like Retrieval-Augmented Generation (RAG) and provides flexibility for building customized AI pipelines. Additionally, "Chains" offers conditional execution to help developers validate outputs before moving forward, optimizing the overall workflow.

Neo4j devhub: https://bit.ly/4ghZuIk