What You'll Learn

• Orchestrate Multi-Agent Systems with LangGraph: Define and manage specialized agents — an EDA Expert, a Statistician, and a Visualization Expert — each handling a distinct stage of a complex machine learning pipeline.
• Integrate LLMs into Data Science Workflows: Connect your application to OpenAI's GPT-4o-mini model and understand how language models can power the reasoning layer of an automated pipeline.
• Build Reusable ML Tools with LangChain: Use the @tool decorator to wrap data science functions as callable agent actions, making your preprocessing and training steps modular and independently testable.
• Manage Shared Agent State: Learn how AgentState and LangGraph's StateGraph allow multiple agents to share data through a common pipeline without tightly coupling their logic.
• Evaluate and Visualize Model Performance: Interpret Mean Squared Error (MSE) and scatter plots to assess how well your Random Forest model predicts CTR, and understand what the results tell you about your data.
• Design Extensible Pipelines: Structure your workflow so that adding a new agent, swapping a model, or inserting a new preprocessing step requires minimal changes to the rest of the system.

Who Should Enroll

• Python Developers looking to expand their skills into Generative AI, agent orchestration, and automated ML pipelines.
• Data Science Beginners who want a hands-on introduction to the full machine learning workflow — preprocessing, training, and evaluation — within a structured, guided project.
• AI Enthusiasts curious about how frameworks like LangGraph can simplify the creation of complex, multi-step AI behaviors without sacrificing clarity or control.
• Software Engineers interested in understanding how agentic design patterns apply to real-world data tasks beyond chatbots and conversational interfaces.
• Students and Educators seeking a practical, applied example of how modern AI tooling integrates with classical machine learning techniques.

Why Enroll

What You'll Need

• Basic Python programming knowledge.
• Familiarity with fundamental data science concepts like features, labels, and train/test splits (helpful but not required — everything is explained as you go).
• Interest in Generative AI, machine learning automation, and modular software design.
• [OPTIONAL] An OpenAI API key for LLM integration (our learning environment provides you an API key for free!)