Teach on Udemy

Turn what you know into an opportunity and reach millions around the world.

Learn More

Your cart is empty.

Keep shopping

14 Days to Building AI Systems & Agents

Name: 14 Days to Building AI Systems & Agents
Rating: 4.6 (17 reviews)

Learn modern AI development by building agents, workflows, and real AI applications.

Created bySchool of AI

Last updated 2/2026

English

English [Auto],

What you'll learn

Design and architect end-to-end AI systems, including UI, backend logic, memory, tools, and automation workflows.
Build and deploy AI agents that can reason, execute actions, and integrate with external tools and APIs.
Implement structured outputs, validation guardrails, and safety checks to make AI systems reliable and production-ready.
Develop stateful AI applications with multi-turn memory and persistent session handling.
Create and automate multi-step AI workflows that chain reasoning, extraction, decision logic, and action execution.
Deploy AI applications securely to the cloud using environment variables and production best practices.
Document, demo, and package AI projects into portfolio-ready assets suitable for job applications or client work.
Apply prompt engineering, evaluation strategies, and regression testing to improve AI system performance over time.

Course content

14 sections • 57 lectures • 4h 41m total length

Certificate of Completion0:29
Enroll in any Udemy course, finish it, download your Udemy certificate, email it to schoolofaillc at gmail.com, and receive the official School of AI certificate after verification.
Foundations: How AI Systems Actually Work9:41
Reset your thinking about ai by mastering the system loop—inputs, processing, model, output, and feedback—to design reliable ai systems rather than chasing buzzwords.
Hands-On Lab: Map a real-world problem into an AI system diagram2:50
Map a real-world problem into an AI system diagram (Description)5:48
Map real-world problems into a simple, repeatable ai system design using the input–processing–model–output–feedback loop, producing a one-page blueprint with success criteria and failure mitigation.
One-page AI system blueprint1:21

Environment Setup for AI Development8:43
Learn how to set up a reproducible AI development environment with Python, virtual environments, a clean project structure, and Git version control to enable smooth, scalable AI projects.
Hands-On Lab: Structure a clean AI project repo2:09
Structure a clean AI project repo (Description)6:03
Master a clean ai project repo by installing python 3.10+, creating a dedicated folder and virtual environment, writing a readme and gitignore, and tracking with git and GitHub.
Working AI dev environment + GitHub repo initialized1:17

Working with APIs & LLMs8:34
Learn to build AI-powered software by calling LLM APIs with Python, control prompts, temperature, and tokens, and deploy a CLI assistant that securely uses environment-based API keys.
Hands-On Lab: Build a simple text-in → text-out app2:02
Build a simple text-in → text-out app (Description)7:16
Learn to implement the text-in, text-out pattern by calling an LLM API from Python, building a secure, command-line AI assistant and validating setup end-to-end.
CLI-based AI assistant script1:08

Prompt Engineering That Actually Works7:40
Turn prompting into engineering by defining system prompts and structured formats with clear roles, tasks, constraints, and guardrails for reliable, reusable ai interactions.
Hands-On Lab: Turn a weak prompt into a robust prompt & Build prompt templates2:57
Turn a weak prompt into a robust prompt & Build prompt templates (Description)7:08
Analyze a weak prompt, diagnose failures, redesign with clear structure and guardrails, and convert it into a reusable template library for reliable AI across use cases.
Prompt library (reusable templates)2:26

Structured Outputs & JSON Responses7:43
Design AI outputs as structured, typed data that machines can consume safely using JSON and schemas. Validate and parse results to prevent failures and enable reliable AI systems.
Hands-On Lab: Force AI to return structured JSON & Parse AI outputs in code3:19
Force AI to return structured JSON & Parse AI outputs in code (Description)5:47
Force AI to return structured JSON and parse the output in code. Define a strict JSON schema, validate required fields, and reject noncompliant results to build reliable AI-enabled systems.
AI function returning validated JSON data1:37

Working with Documents (RAG Basics)7:00
Explore retrieval augmented generation (rag) to ground answers in your documents with embeddings, a vector store, and a two-step retrieve-then-generate process.
Hands-On Lab: Load documents; Embed them; Ask questions over your own data4:38
Load documents; Embed them; Ask questions over your own data (Description)5:43
Build an end-to-end retrieval pipeline that ingests documents, creates embeddings, retrieves relevant context, and generates grounded answers with no memory or guessing.
Document-aware AI assistant1:36

Building a Simple AI Web App6:08
Wrap your ai logic behind a simple web interface to turn code into a usable product. Create a minimal back-end api and a basic front-end to enable browser-based interaction.
Hands-On Lab: Build a simple web UI; Connect it to their AI backend2:58
Build a simple web UI; Connect it to their AI backend (Description)4:55
Connect a simple web UI to your existing AI backend to turn input into an end-to-end AI product. Build a minimal, complete system with a front-end, back-end, and API.
Working AI web app (local)1:57

Memory & State in AI Apps5:51
Implement memory and state management to persist context across turns, using session IDs, storing conversation turns, and replaying context to create a coherent, multi-turn ai app.
Hands-On Lab: Add conversation memory; Persist user sessions3:58
Add conversation memory; Persist user sessions (Description)5:15
Move from single-turn chats to real conversations by designing state with session IDs. Create a session-based memory layer that stores, replays, and persists conversation history for coherent interactions.
Multi-turn conversational AI app3:38

AI Agents: From Chatbots to Doers6:03
Move from chatbots to goal-driven ai agents that reason, plan, select tools, execute actions, observe results, and iterate toward completion within bounded autonomy.
Hands-On Lab: Build a task-executing AI agent ; Give it tools5:28
Build a task-executing AI agent ; Give it tools (Description)5:59
Move from conversational ai to operational ai by building a bounded think-act-observe loop, a tool registry, and tools that create files, search documents, and perform calculations.
AI agent that completes real tasks3:54

Automation & Workflows6:14
Learn to design deterministic, trigger-based workflows that orchestrate multiple ai steps, chaining outputs from one stage to the next while balancing automation with human oversight.
Hands-On Lab: Build a multi-step AI workflow; Automate a real use case3:51
Build a multi-step AI workflow; Automate a real use case (Description)9:14
Design and implement a real end-to-end automation pipeline that processes inputs from a folder trigger, chains six ai-powered stages, and logs results.
End-to-end AI workflow1:37

Requirements

Basic familiarity with Python (variables, functions, and installing packages with pip).
Basic familiarity with Python (variables, functions, and installing packages with pip).
Ability to run commands in a terminal (basic command-line usage).
Ability to run commands in a terminal (basic command-line usage).
An internet connection for accessing AI APIs and cloud deployment platforms.
An internet connection for accessing AI APIs and cloud deployment platforms.

Description

“This course contains the use of artificial intelligence”

14 Days to Building AI Systems & Agents is a hands-on, implementation-focused program designed to teach you how to build real, production-ready AI applications from the ground up. This is not a theory-heavy course. It is a practical, structured journey into AI system architecture, LLM integration, AI agents, and automation workflows.

Over 14 days, you will move beyond simple prompt experimentation and learn how modern AI development actually works. You will design full-stack AI systems that combine frontend interfaces, backend APIs, multi-turn memory, tool execution, and validation guardrails. Instead of building isolated scripts, you will build cohesive AI products.

The course begins with core foundations: understanding how AI systems, machine learning models, and large language models (LLMs) fit together. You will learn how to work with LLM APIs, manage tokens and temperature, and implement structured prompts that produce reliable outputs. From there, you will progress into building stateful systems with persistent memory, enabling your AI applications to handle multi-turn conversations and maintain context.

As the course advances, you will implement structured JSON outputs, enforce schema validation, and apply AI safety guardrails to make your systems dependable. You will also build practical AI agents that can execute tools, interact with files, and perform multi-step reasoning. These agents will not just respond—they will act.

You will then learn how to design and implement automation workflows, chaining multiple AI steps into deterministic pipelines. By combining reasoning, tools, and validation, you will create end-to-end AI workflows capable of solving real operational tasks.

Deployment is a core focus. You will package your system, configure environment variables securely, and deploy your AI application to the cloud. You will learn how to move from a local prototype to a live, shareable product.

Finally, you will transform your capstone into a portfolio-ready AI project. You will write professional documentation, create architecture diagrams, record a demo, and polish your GitHub repository. By the end of the course, you will have a complete AI product demonstrating skills in AI system design, agent architecture, workflow automation, API integration, and cloud deployment.

This course is ideal for developers, product builders, and technical professionals who want practical experience building modern AI systems. No advanced math is required. You only need basic Python knowledge and a willingness to build.

If you want to go beyond prompts and learn how to engineer real AI-powered applications, integrate AI agents, and deploy reliable AI workflows, this 14-day program gives you the structure, clarity, and hands-on experience to do exactly that.

Build. Integrate. Deploy. Ship real AI systems.

Who this course is for:

Developers who want to move beyond basic prompts and learn how to build complete AI systems and agents.
Software engineers interested in integrating LLMs, memory, tools, and automation into real-world applications.
Product builders and startup founders who want to prototype and launch AI-powered applications quickly.
Technical professionals looking to transition into AI engineering or agent-based system development.
Backend or full-stack developers who want hands-on experience deploying production-ready AI applications.
AI enthusiasts who have experimented with ChatGPT or APIs and now want to understand system architecture and integration.
Freelancers and consultants who want to offer AI automation and agent-building services to clients.
Students and career switchers aiming to build a strong AI portfolio with deployable projects.

14 Days to Building AI Systems & Agents

What you'll learn

Explore related topics

Course content

Day 1 – Foundations: How AI Systems Actually Work5 lectures • 20min

Day 2 – Environment Setup for AI Development4 lectures • 18min

Day 3 – Working with APIs & LLMs4 lectures • 19min

Day 4 – Prompt Engineering That Actually Works4 lectures • 20min

Day 5 – Structured Outputs & JSON Responses4 lectures • 18min

Day 6 – Working with Documents (RAG Basics)4 lectures • 19min

Day 7 – Building a Simple AI Web App4 lectures • 16min

Day 8 – Memory & State in AI Apps4 lectures • 19min

Day 9 – AI Agents: From Chatbots to Doers4 lectures • 21min

Day 10 – Automation & Workflows4 lectures • 21min

Requirements

Description

Who this course is for: