explaingit

microsoft/rd-agent

12,993PythonAudience · researcherComplexity · 4/5Setup · hard

TLDR

A Microsoft AI system that automates the research-and-development loop in data science and quantitative finance, it proposes ideas, writes and runs code, evaluates results, and iterates without human involvement.

Mindmap

mindmap
  root((rd-agent))
    What it does
      AI R and D automation
      Research-code loop
      Autonomous experiments
    Scenarios
      Quantitative finance
      Kaggle competitions
      LLM fine-tuning
    Architecture
      Research agent
      Development agent
      Web interface
    Audience
      Data scientists
      ML researchers
Click or tap to explore — scroll the page freely

Code map

Detail Auto

An interactive map of this repo's files and how they connect — its source is parsed live in your browser. Click Visualize to build it.

filefunction / class

Things people build with this

USE CASE 1

Automatically generate and backtest quantitative trading signals by having the agent propose formulas, code them, run historical backtests, and refine the approach

USE CASE 2

Enter Kaggle machine learning competitions by letting the agent read the problem, propose solutions, write and debug code, and submit results autonomously

USE CASE 3

Fine-tune large language models without writing training recipes by hand, the agent designs experiments and runs them autonomously

USE CASE 4

Assign a powerful reasoning model to the research role and a cheaper model to the coding role to balance quality against cost

Tech stack

Python

Getting it running

Difficulty · hard Time to first run · 1day+

Requires API keys for one or more AI providers, GPU resources are needed for self-hosted LLM fine-tuning experiments.

In plain English

RD-Agent is a project from Microsoft that uses AI to automate parts of research and development work, specifically in data science and machine learning. The core idea is that a lot of R&D work follows repeatable patterns: propose an idea, write code to test it, evaluate the results, and repeat. RD-Agent builds software agents that can carry out this cycle with minimal human involvement. The project targets several practical scenarios. In quantitative finance, it can automatically generate and test trading signals by proposing mathematical formulas, coding them up, running backtests against historical market data, and refining the approach based on results. In general data science competitions, it can enter Kaggle-style machine learning contests by reading the problem description, proposing solutions, writing and debugging code, and submitting results. It currently ranks at the top of MLE-bench, a public benchmark that measures how well AI agents handle machine learning engineering tasks drawn from 75 real Kaggle competitions. The system is built around a loop of two main roles: a research agent that proposes ideas and a development agent that implements them. Different AI models can be assigned to each role. For example, one configuration uses a more capable reasoning model for the research side and a faster, cheaper model for writing code on the development side. Installation is through the standard Python package manager. The project supports multiple AI providers through a compatibility layer, so you are not locked into a single service. A web interface is included for watching runs in real time and reviewing the agent's reasoning traces. The repository also covers fine-tuning large language models autonomously, where the agent designs training recipes and runs experiments without human intervention. Papers from this project have been accepted at major machine learning conferences including NeurIPS 2025 and ICML 2026. The full README is longer than what was shown.

Copy-paste prompts

Prompt 1
Set up RD-Agent to automatically propose and backtest momentum trading signals on a historical stock price dataset
Prompt 2
How do I configure RD-Agent to use one AI model for research proposals and a different, cheaper model for writing the implementation code?
Prompt 3
Run RD-Agent on a Kaggle tabular classification competition and have it submit its best solution automatically
Prompt 4
Use RD-Agent's web interface to watch a live agent run and review the reasoning traces the agent produced step by step
Prompt 5
How do I add a new AI provider to RD-Agent using its compatibility layer so I am not locked into a single service?
Open on GitHub → Explain another repo

← microsoft on gitmyhub — every repo by this author, as a profile.

Verify against the repo before relying on details.