What is RD-Agent used for?

RD-Agent is Microsoft's open-source framework that uses LLM agents to automate data-driven research and development, including data science, Kaggle competitions, quant trading and fine-tuning. It is a good fit for automating machine learning experiments, researching quantitative trading strategies and working on Kaggle-style competitions.

Is RD-Agent open source?

Yes. RD-Agent is open source under the MIT licence. Its source code is on GitHub at microsoft/RD-Agent and is written mainly in Python.

Yes. RD-Agent is open source, so the software itself is free to use.

What are some alternatives to RD-Agent?

Similar open-source tools in the AI Infrastructure category include TPOT, Langflow and InternVL. SaaS products in the same category include Amazon Bedrock, Azure AI Foundry and Azure Machine Learning.

Is RD-Agent actively maintained?

Yes. The most recent commit to RD-Agent was on 30 September 2026, and the latest release is v1.0.0, published on 23 September 2026. The project has 15k stars on GitHub.

RD-Agent: open-source AI infrastructure

About RD-Agent

R&D-Agent is a Python framework from Microsoft that aims to automate the repetitive parts of research and development in data and AI work. It targets the data and model side of R&D, using large language model agents to drive experiments end to end so that researchers and data scientists can spend more time reviewing results.

The project covers several scenarios, including a general data science agent, a Kaggle agent, an application to quantitative trading, and an agent for LLM fine-tuning. LiteLLM is the default backend, which lets it work with multiple LLM providers, and a web interface started with the rdagent command offers real-time interaction and trace viewing for most scenarios.

The code is MIT licensed and runs as a Python package that you install and operate yourself. The repository is accompanied by a live demo, a demo video, documentation, a technical report and several research papers, which makes it easier to evaluate the approach before adopting it.

Key features

LLM agents that automate data science experiments
Kaggle competition scenario agent
Quantitative trading research agent
LLM fine-tuning agent
LiteLLM backend for many model providers
Web UI with real-time trace viewing

Good fit for

→Automating machine learning experiments
→Researching quantitative trading strategies
→Working on Kaggle-style competitions

Built with: Python
Tags: ai-agents; llm; automation; data-science; machine-learning; kaggle; research; python; microsoft

RD-Agent: questions and answers

What is RD-Agent used for?: RD-Agent is Microsoft's open-source framework that uses LLM agents to automate data-driven research and development, including data science, Kaggle competitions, quant trading and fine-tuning. It is a good fit for automating machine learning experiments, researching quantitative trading strategies and working on Kaggle-style competitions.
Is RD-Agent open source?: Yes. RD-Agent is open source under the MIT licence. Its source code is on GitHub at microsoft/RD-Agent and is written mainly in Python.
Is RD-Agent free?: Yes. RD-Agent is open source, so the software itself is free to use.
What are some alternatives to RD-Agent?: Similar open-source tools in the AI Infrastructure category include TPOT, Langflow and InternVL. SaaS products in the same category include Amazon Bedrock, Azure AI Foundry and Azure Machine Learning.
Is RD-Agent actively maintained?: Yes. The most recent commit to RD-Agent was on 30 September 2026, and the latest release is v1.0.0, published on 23 September 2026. The project has 15k stars on GitHub.