Source: https://seerepo.com/r/microsoft/RD-Agent/built
Markdown source: https://seerepo.com/r/microsoft/RD-Agent/built.md
Description: RD-Agent is a Python repo built with Flask, werkzeug, flask-cors, Streamlit, Vite, and 2 more.

# What is RD-Agent built with?

RD-Agent is a Python repo built with Flask, werkzeug, flask-cors, Streamlit, Vite, and 2 more. It uses AI through accelerate, datasets, gym, peft, pydantic-ai-slim, and 13 more, makes its content with beautifulsoup4, myst-parser, pypdf, nbformat, pymupdf, and 6 more, validates its data with genson and Pydantic, is tested with Pytest, Selenium, and webdriver-manager, deploys with docker, loguru, and Prefect, and uses pyarrow, tables, Hugging Face, and Kaggle. It has 83,518 written lines, with 41 test files.

Read on November 8, 2026, at commit 484776c.

Repository: https://github.com/microsoft/RD-Agent

## Frontend (/)

### Styles

- **furo**
- **sphinx-rtd-theme**

### Interface

- **[seaborn](https://seerepo.com/uses/seaborn)**
- **[Plotly](https://seerepo.com/uses/plotly)**: 8 files
- **[Rich](https://seerepo.com/uses/rich)**: 3 files
- **[matplotlib](https://seerepo.com/uses/matplotlib)**: 2 files

## Backend (/)

### The base

- **[Flask](https://seerepo.com/uses/flask)**: 3 files
- **[werkzeug](https://seerepo.com/uses/werkzeug)**: 2 files

### Data

- **[pyarrow](https://seerepo.com/uses/pyarrow)**
- **tables**

### AI

- **[accelerate](https://seerepo.com/uses/accelerate)**
- **[datasets](https://seerepo.com/uses/datasets)**
- **gym**
- **[peft](https://seerepo.com/uses/peft)**
- **pydantic-ai-slim**
- **pyserini**
- **[rank-bm25](https://seerepo.com/uses/rank-bm25)**
- **[spaCy](https://seerepo.com/uses/spacy)**
- **trl**
- **vLLM**
- **[scikit-learn](https://seerepo.com/uses/scikit-learn)**: 66 files
- **[litellm](https://seerepo.com/uses/litellm)**: 6 files
- **[torch](https://seerepo.com/uses/torch)**: 5 files
- **MLflow**: 3 files
- **[tiktoken](https://seerepo.com/uses/tiktoken)**: 2 files
- **[LangChain](https://seerepo.com/uses/langchain)**: 1 file
- **[transformers](https://seerepo.com/uses/transformers)**: 1 file

### Services

- **[OpenAI](https://seerepo.com/uses/openai)**: 2 files · [rdagent/oai/backend/base.py, line 30](https://github.com/microsoft/RD-Agent/blob/484776c211e4fbbeef03e0ec00d6bbee7362a4f4/rdagent/oai/backend/base.py#L30)
- **[Hugging Face](https://seerepo.com/uses/hugging-face)**: 1 file · [rdagent/scenarios/finetune/download/hf.py, line 49](https://github.com/microsoft/RD-Agent/blob/484776c211e4fbbeef03e0ec00d6bbee7362a4f4/rdagent/scenarios/finetune/download/hf.py#L49)
- **Kaggle**: 1 file · [rdagent/scenarios/kaggle/kaggle_crawler.py, line 198](https://github.com/microsoft/RD-Agent/blob/484776c211e4fbbeef03e0ec00d6bbee7362a4f4/rdagent/scenarios/kaggle/kaggle_crawler.py#L198)

## Support (/)

### The base

- **[flask-cors](https://seerepo.com/uses/flask-cors)**
- **[Streamlit](https://seerepo.com/uses/streamlit)**: 18 files

### Content

- **[beautifulsoup4](https://seerepo.com/uses/beautifulsoup4)**
- **myst-parser**
- **[pypdf](https://seerepo.com/uses/pypdf)**
- **nbformat**: 3 files
- **[pymupdf](https://seerepo.com/uses/pymupdf)**: 1 file

### Validation

- **genson**
- **[Pydantic](https://seerepo.com/uses/pydantic)**: 10 files

### Testing

- **Pytest**: 12 files
- **Selenium**: 1 file
- **webdriver-manager**: 1 file

### Deploy

- **[docker](https://seerepo.com/uses/docker)**: 3 files
- **[loguru](https://seerepo.com/uses/loguru)**: 3 files
- **Prefect**: 1 file

### Tools

- **autodoc-pydantic**
- **Black**
- **build**
- **coverage**
- **isort**
- **mypy**
- **Ruff**
- **Setuptools**
- **Sphinx**
- **sphinx-autobuild**
- **sphinx-click**
- **sphinx-togglebutton**
- **tensorboard**
- **twine**
- **types-psutil**
- **types-pytz**
- **types-pyyaml**
- **types-tqdm**
- **wheel**
- **fire**: 13 files
- **Typer**: 7 files
- **setuptools-scm**: 1 file
- **tree-sitter**: 1 file
- **tree-sitter-python**: 1 file

### Other

- **duckduckgo-search**
- **gdown**
- **python-levenshtein**
- **snowballstemmer**
- **thefuzz**
- **pandas**: 223 files
- **NumPy**: 56 files
- **pydantic-settings**: 15 files
- **requests**: 8 files
- **tqdm**: 8 files
- **pyyaml**: 7 files
- **filelock**: 4 files
- **python-dotenv**: 2 files
- **randomname**: 2 files
- **fuzzywuzzy**: 1 file
- **humanize**: 1 file
- **nest-asyncio**: 1 file
- **networkx**: 1 file

## Frontend (web)

### The base

- **[Vite](https://seerepo.com/uses/vite)**: [in web/package.json](https://github.com/microsoft/RD-Agent/blob/484776c211e4fbbeef03e0ec00d6bbee7362a4f4/web/package.json)
- **[Vue](https://seerepo.com/uses/vue)**: 26 files · [in web/package.json](https://github.com/microsoft/RD-Agent/blob/484776c211e4fbbeef03e0ec00d6bbee7362a4f4/web/package.json)
- **[vue-router](https://seerepo.com/uses/vue-router)**: 4 files

### Styles

- **github-markdown-css**: 1 file

### Interface

- **[Element](https://seerepo.com/uses/element)**: 5 files
- **[echarts](https://seerepo.com/uses/echarts)**: 2 files

## Support (web)

### Content

- **[highlight.js](https://seerepo.com/uses/highlightjs)**: 2 files
- **[katex](https://seerepo.com/uses/katex)**: 2 files
- **[marked](https://seerepo.com/uses/marked)**: 2 files
- **[markdown-it](https://seerepo.com/uses/markdown-it)**: 1 file
- **markdown-it-texmath**: 1 file
- **[prismjs](https://seerepo.com/uses/prismjs)**: 1 file

### Other

- **Axios**: 1 file
- **crypto-js**: 1 file
- **jQuery**: 1 file
- **jszip**: 1 file

## Other repos with a similar stack

- [datawhalechina/hello-agents](https://seerepo.com/r/datawhalechina/hello-agents/built): Flask, OpenAI, Hugging Face, Streamlit, Vite, Vue, vue-router, Plotly, Rich, matplotlib, accelerate, datasets, peft, rank-bm25, spaCy, scikit-learn, torch, tiktoken, transformers, beautifulsoup4, pypdf, pymupdf, Pydantic, loguru, Element, echarts, highlight.js, katex, marked, markdown-it · [Compare](https://seerepo.com/compare/datawhalechina/hello-agents/vs/microsoft/RD-Agent)
- [bojieli/ai-agent-book](https://seerepo.com/r/bojieli/ai-agent-book/built): Flask, pyarrow, OpenAI, Hugging Face, Vite, seaborn, Plotly, Rich, matplotlib, accelerate, datasets, peft, rank-bm25, trl, vLLM, scikit-learn, litellm, torch, tiktoken, transformers, beautifulsoup4, pypdf, pymupdf, Pydantic, loguru, katex · [Compare](https://seerepo.com/compare/bojieli/ai-agent-book/vs/microsoft/RD-Agent)
- [ray-project/ray](https://seerepo.com/r/ray-project/ray/built): Flask, werkzeug, pyarrow, Hugging Face, flask-cors, Plotly, Rich, matplotlib, accelerate, datasets, gym, peft, scikit-learn, torch, MLflow, tiktoken, LangChain, transformers, beautifulsoup4, nbformat, Pydantic, docker · [Compare](https://seerepo.com/compare/microsoft/RD-Agent/vs/ray-project/ray)
- [run-llama/llama_index](https://seerepo.com/r/run-llama/llama_index/built): pyarrow, OpenAI, Rich, matplotlib, datasets, peft, vLLM, scikit-learn, litellm, torch, tiktoken, LangChain, transformers, beautifulsoup4, pypdf, nbformat, pymupdf, Pydantic, docker · [Compare](https://seerepo.com/compare/microsoft/RD-Agent/vs/run-llama/llama_index)
- [unslothai/unsloth](https://seerepo.com/r/unslothai/unsloth/built): OpenAI, Hugging Face, Vite, Rich, matplotlib, accelerate, datasets, peft, scikit-learn, torch, tiktoken, transformers, pymupdf, Pydantic, loguru, katex, marked · [Compare](https://seerepo.com/compare/microsoft/RD-Agent/vs/unslothai/unsloth)
- [gradio-app/gradio](https://seerepo.com/r/gradio-app/gradio/built): OpenAI, Hugging Face, Vite, Plotly, matplotlib, accelerate, datasets, spaCy, scikit-learn, torch, LangChain, transformers, pymupdf, Pydantic, katex, marked, prismjs · [Compare](https://seerepo.com/compare/gradio-app/gradio/vs/microsoft/RD-Agent)

## Frequently asked questions

### What language is RD-Agent written in?

RD-Agent is written in Python, with 83,518 written lines.

### What framework does RD-Agent use?

RD-Agent is built with Flask, werkzeug, flask-cors, Streamlit, Vite, Vue, and vue-router.

### Does RD-Agent use AI?

Yes. RD-Agent uses AI through accelerate, datasets, gym, peft, pydantic-ai-slim, pyserini, rank-bm25, spaCy, trl, vLLM, scikit-learn, litellm, torch, MLflow, tiktoken, LangChain, transformers, and OpenAI.

### What services does RD-Agent use?

RD-Agent uses pyarrow, tables, Hugging Face, and Kaggle.

### Does RD-Agent have tests?

Yes. RD-Agent has 41 test files and is tested with Pytest, Selenium, and webdriver-manager.

### Which repos have a stack like RD-Agent's?

The ones that share most of its stack: datawhalechina/hello-agents, bojieli/ai-agent-book, and ray-project/ray.

## What is your repo built with?

Any public repo, with no account: https://seerepo.com/r/owner/repo/built, with the owner and name it has on GitHub.