Your cart is empty.
Keep shopping7 Days Certified AI Agents with Python: Autonomous Apps
New
302 students
7 Days Certified AI Agents with Python: Autonomous Apps
Build autonomous AI agents with Python, Ollama, tools, memory, RAG, research, and multi-agent workflows
Created bySchool of AI, Arjun Vaid
Last updated 7/2026
English
What you'll learn
- Understand how AI agents differ from traditional chatbots and standard LLM applications.
- Build a personal AI assistant in Python using Ollama and Streamlit.
- Connect Python applications to local language models through Ollama.
- Design effective system prompts, agent roles, and instruction patterns.
- Create structured outputs using Pydantic and JSON schemas.
- Build agents that can break complex goals into smaller, actionable tasks.
- Implement planner, executor, analyst, writer, and reviewer agent patterns.
- Create tool-using agents that can call Python functions, APIs, databases, and file utilities.
- Build reusable tools for calculations, search, file handling, and data analysis.
- Add short-term conversation memory and persistent user preferences.
- Process PDF documents and build Retrieval-Augmented Generation workflows.
- Generate embeddings with Ollama and store them in a vector database.
- Build an autonomous research agent that searches, collects, evaluates, and summarizes information.
- Design multi-agent workflows where specialized agents collaborate on a shared goal.
- Orchestrate agent workflows using LangGraph and shared state.
- Add human approval checkpoints before important actions are executed.
- Implement reviewer loops that evaluate and improve agent-generated results.
- Add logging, retries, error handling, validation, and workflow safeguards.
- Build a complete airline disruption assistant that combines planning, tools, memory, RAG, research, and multiple agents.
- Deploy a complete autonomous AI application with a Streamlit interface.
Explore related topics
Course content
8 sections • 119 lectures • 13h 11m total length
Requirements
- Basic knowledge of Python, including variables, functions, lists, dictionaries, and classes.
- A computer running Windows, macOS, or Linux.
- Python 3.11 or later installed on your computer.
- Basic familiarity with installing Python packages using pip.
- A code editor such as Visual Studio Code, PyCharm, or another Python-compatible editor.
- Ollama installed locally for running language and embedding models.
- At least 8 GB of RAM is recommended; 16 GB or more provides a better experience with larger local models.
- An internet connection is helpful for installing packages, downloading Ollama models, and completing optional web-research exercises.
- No previous experience with AI agents, LangGraph, vector databases, or Retrieval-Augmented Generation is required.
- No paid AI API subscription is required because the course uses local models through Ollama.
- Beginners with basic Python knowledge can follow the course because each system is built step by step.
Description
Learn how to build practical AI agents with Python using modern tools such as Ollama, LangGraph, Streamlit, Pydantic, ChromaDB, and Retrieval-Augmented Generation.
In this hands-on course, you will explore how agentic AI applications are designed, developed, and tested through real-world project examples. Instead of focusing only on basic chatbot interactions, this course introduces structured AI agent workflows that can use tools, retrieve information, follow instructions, and support human decision-making.
You will begin with the foundations of AI agent development, including how agents receive goals, manage prompts, produce structured outputs, and work with Python-based logic. You will build a personal AI assistant project and learn how to organize inputs, outputs, conversation history, and response formats.
Next, you will practice AI planning and task decomposition. You will learn how to break larger objectives into smaller steps, design reliable JSON outputs, and use Pydantic models to validate structured responses. These skills are useful for creating more predictable and maintainable Generative AI applications.
The course then introduces tool-using AI agents. You will create Python functions that an agent can call for tasks such as calculations, file reading, API interactions, data lookup, and controlled automation. You will also explore important engineering concepts such as tool permissions, validation, retries, error handling, and human oversight.
A major part of the course focuses on RAG with Python. You will learn how to process PDF documents, split text into chunks, create embeddings with Ollama, store data in ChromaDB, and retrieve relevant content using semantic search. These lessons demonstrate how Retrieval-Augmented Generation can help AI systems work with private documents and trusted knowledge sources.
You will also build an autonomous research agent project that demonstrates query generation, information gathering, summarization, gap identification, and structured report creation. You will learn how autonomous loops work and how to design controls that help keep AI workflows focused and manageable.
Later in the course, you will explore multi-agent systems using LangGraph. You will design specialized agents such as a planner, researcher, analyst, writer, reviewer, and supervisor. These agents will communicate through shared state and demonstrate how collaborative AI workflows can be orchestrated in a structured way.
Throughout the course, you will work with a simulated airline disruption example. This project shows how an AI workflow can review booking details, compare possible flight options, reference policy information, generate recommendations, and request human approval before any final action.
By the end of the course, you should have hands-on experience with Python AI agents, local LLMs, Ollama AI development, LangGraph workflows, vector databases, AI memory, RAG systems, tool calling, human-in-the-loop AI, and multi-agent collaboration.
This course is ideal for Python developers, AI beginners, automation specialists, software engineers, data professionals, entrepreneurs, and anyone interested in building practical local-first AI agent projects.
Who this course is for:
- Python developers who want to move beyond basic chatbots and build autonomous AI applications.
- Beginners in generative AI who already understand basic Python and want a practical introduction to AI agents.
- Software developers who want to learn planning, tool calling, memory, RAG, and multi-agent orchestration.
- Data scientists and machine learning practitioners who want to build interactive agent-based applications.
- Automation developers who want AI systems that can perform tasks, use tools, and make controlled decisions.
- Backend developers interested in integrating local language models into Python applications.
- Students and professionals who want hands-on experience with Ollama, Streamlit, Pydantic, Chroma, and LangGraph.
- Developers who want to build private, local-first AI applications without relying on paid cloud-based LLM APIs.
- Entrepreneurs and product builders who want to prototype AI assistants, research agents, workflow agents, and multi-agent systems.
- Technical instructors and course creators looking for practical AI-agent projects they can demonstrate or adapt.
- Anyone who wants to understand how complete autonomous workflows are designed, tested, reviewed, and deployed.
- Learners who prefer project-based teaching and want to finish the course with multiple portfolio-ready Python applications.