Burr can be used to power a variety of applications, including simple gpt-like chatbots, stateful RAG-based chatbots, machine learning pipelines, and simulations. It comes with a host of integrations, including tooling to build a UI in streamlit and watch your state machine execute. Burr is named after Aaron Burr, the third Vice President of the United States, and was originally built as a harness to handle state between executions of Hamilton DAGs. The tool is open for contributions and has a roadmap of features and improvements planned for the future.
Key takeaways:
- Burr is a tool that simplifies the development of applications that make decisions based on state, such as chatbots, agents, and simulations. It includes a UI for real-time tracking and monitoring of these decisions.
- Burr allows you to express your application as a state machine, making it useful for a wide range of contexts, from chatbots to machine learning workflows and complex forecasting simulations.
- While Burr provides a host of integrations and tools, it does not dictate how to build your models, query APIs, or manage your data. Instead, it helps tie these elements together in a scalable and logical way.
- Burr is named after Aaron Burr, the third Vice President of the United States, and was originally built as a harness to handle state between executions of Hamilton DAGs.