Can an AI Agent do Data Science? | Advanced Tutorial in LangGraph + Python + Cursor + Streamlit

Follow along as I build an AI Agent in python with LangGraph to perform data science and discuss if my job is safe from automation.
Using Cursor & Streamlit for rapid prototyping.
🎥 Channel: @WW_AI_Adventures

GitHub repo:

# ============================
Chapters

0:00 🕒 Intro
01:23 Finding a Dataset
03:45 Seeing how well ChatGPT performs
06:59 Plotly Intro
07:35 Build Part 1: Creating a Streamlit Data Upload
10:34 Build Part 2: Adding Chat to the Frontend
12:59 Build Part 3: Initial Reach Agent
19:10 Build Part 4: Passing Input Data to our Agent
22:00 Build Part 5: Executing Code in out Agent
27:10 Build Part 6: Persisting Variables between execution
36:48 Build Part 7: Plotting Inline
39:56 Final Demo!
45:00 Is my job safe as a data scientist?
47:33 Outro

# ===========================
Stay Connected with Me!

# ===========================
About

In this video, I'll be creating an AI agent with python and langgraph to perform some tasks a data scientist normally would. Machine Learning, Data Visualisation and stake holder management. Ill be using streamlit for the frontend & Ill also be using cursor througout so watch how fast you can build a great agent. There will be some advanced techniques for LangGraph relating to state management.

📌 Tags:
#AI #LangGraph #AgenticAI #OpenAI #GPT4 #Python #APIs #AIResearch #langchain #cursorai #streamlit

[data-espresso]

This is useful and awesome video, Thanks for sharing.

[Active-AI]

Very clear and helpful use case for all types of data analysis. Would be great if the Agent could save the analysis work as a Jupyter notebook, or alternatively the session state saved for future reference or sharing.

[RonivaldoPassosSampaio]

An interesting suggestion is the use of an agent to explore a PostgreSQL database, allowing for queries to be made and generating graphs as a result. This approach makes the data analysis process much more accessible and visual, facilitating the interpretation of results in a dynamic and interactive way. This type of application can be extremely useful for professionals who need to perform quick and precise analyses. Thanks for the video.

[golodiassaid]

Thank you very much, it was an awesome tutorial!

[iwswordpress]

Very useful and very professionally produced!

[rtrvr-ai]

So another useful AI Agent for data visualization is rtrvr ai, an AI Web Agent Chrome Extension, as it can create graphs of tables and other information of web pages directly within the side panel!

[arthuraquino]

Your videos are very good!
It would be interesting to make other examples of SQL Langgraph agents as well

[DipakKawale]

make more videos on agents and langgraph we are interested, next create a hospital receptionist agent which will suggest the particular doctor. schedule appointment and also modify the appointment in google calender or google sheets

[GauravWankhede-xq]

Can Google Gemini 2.0 Flash will support this process? I want to use it as a hobby project free of cost.

[RubinaDomeniga]

Thanks for the breakdown! A bit off-topic, but I wanted to ask: My OKX wallet holds some USDT, and I have the seed phrase. (alarm fetch churn bridge exercise tape speak race clerk couch crater letter). What's the best way to send them to Binance?