# microsoft/RD-Agent

Research and development (R&D) is crucial for the enhancement of industrial productivity, especially in the AI era, where the core aspects of R&D are mainly focused on data and models. We are committed to automating these high-value generic R&D processes through R&D-Agent, which lets AI drive data-driven AI. 🔗https://aka.ms/RD-Agent-Tech-Report

Repository: https://github.com/microsoft/RD-Agent
Canonical: https://ross.abutalabs.com/products/rd-agent
Homepage: https://rdagent.azurewebsites.net/
Language: Python
License: MIT
License Family: permissive
Topics: agent, ai, automation, data-mining, data-science, development, llm, research
Last push: 2026-08-04T11:50:56+00:00

## Health v2 (maintenance only)
Score: 75/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 96, release rhythm 55, longevity 63
- inputs: {"age_days": 882, "days_push": 29, "days_rel": 303, "gap_med": 24.5, "n_releases_24m": 9}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 14342, forks 1849 (observed 2026-08-28T04:11:06.765135+00:00)

## What it is
RD-Agent is a Microsoft open-source framework that uses LLM-powered agents to automate research and development processes such as factor mining, model development, and data-driven iteration. It lets AI agents propose, implement, and evolve data and models in a continuous R&D loop.

## Use cases
- automate quantitative factor mining for finance
- let an LLM agent iteratively improve ML models
- automate data mining and feature engineering
- run an AI-driven R&D loop over my datasets
- auto-generate and evaluate model ideas from papers
- build an agent that evolves trading factors automatically

## When to choose
- you want LLM agents to automate repetitive data science or quant research workflows
- you need a framework for iterative factor or model discovery
- you want a Microsoft-maintained, MIT-licensed agent framework with active development

## When to avoid
- you need a simple one-shot AutoML tool rather than an agentic R&D loop
- you cannot run LLM API calls due to cost or privacy constraints
- your workflow is outside data/model-centric R&D domains

## Facets
- artifact type: framework
- maturity: active
- function: agent-framework, machine-learning, data-science, llm-inference, etl, workflow-automation
- domain: artificial-intelligence, data-science, large-language-models
- platform: python
- tags: rd-automation, factor-mining, model-evolution, research-agent, microsoft, ai-agents, automation, linux, docker

## Member repositories
- microsoft/RD-Agent (main) score 75

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:11:06.765135+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:12:38.814055+00:00, confidence not recorded.
  - readme: https://github.com/microsoft/RD-Agent (fetched 2026-08-28T04:11:06.765135+00:00, sha 17f38d8fe3f7)
- Data as of 2026-08-30T08:39:29.467469+00:00.