₹0 for students · Certificate · Beginner

Generative AI for Beginners

21 lessons: build real applications with large language models

Start free in the app

21 lessons · ~15 hours · 30-question final exam · pass mark 70%

About the course

What you’ll learn

Everything you need to start building with generative AI: how LLMs work and how to choose one, using them responsibly, prompt engineering, text, chat, search and image apps, function calling, UX and security for AI apps, the LLMOps lifecycle, retrieval-augmented generation with vector databases, open-source and small models, agents and fine-tuning. Each lesson has a study guide, the full lesson with code samples, and a quiz. Pass every quiz and the final exam to earn a verifiable certificate.

Generative AILLMsPrompt engineeringRAGEmbeddingsFunction callingVector databasesAI agentsFine-tuningResponsible AI

Part of the career track: GenAI and LLM Builder

How the certificate works

  1. Read each lesson’s study guide, then the full lesson and any notebooks or exercises.
  2. Pass every lesson quiz with 70% or more.
  3. Pass the final exam: 30 questions in 45 minutes.
  4. Get a certificate with a public verification link, and add it to LinkedIn.

Syllabus

21 lessons in 5 parts

Foundations of generative AI

1
Introduction to Generative AI and Large Language Models
40 min · Generative AI creates new content; LLMs drive the current wave.
2
Exploring and Comparing Different LLMs
45 min · Choose models by task, cost, latency, privacy and deployment.
3
Using Generative AI Responsibly
40 min · Risks: hallucinations, harmful content, unfairness and over-trust.

Prompting

4
Prompt Engineering Fundamentals
50 min · Prompt engineering designs and iterates prompts for reliable outputs.
5
Creating Advanced Prompts
45 min · Few-shot and chain-of-thought prompting improve multi-step reasoning.

Building applications

6
Building Text Generation Applications
50 min · Use a client library (e.g. OpenAI SDK) with endpoint and key from environment variables.
7
Building Chat Applications
45 min · AI chat apps handle open-ended requests, unlike scripted chatbots.
8
Building Search Applications with Embeddings
45 min · Keyword search matches words; semantic search matches meaning.
9
Building Image Generation Applications
40 min · Diffusion models generate images by gradually denoising random noise, guided by text.
10
Building Low-Code AI Applications
40 min · Low-code platforms build apps and workflows visually with little code.
11
Integrating with Function Calling
45 min · Prompted JSON is inconsistent; function calling returns structured calls.

Shipping AI apps safely

12
Designing UX for AI Applications
35 min · Good AI UX is useful, reliable, accessible and pleasant.
13
Securing Generative AI Applications
45 min · OWASP Top 10 for LLMs: prompt injection, insecure output handling, poisoning, disclosure, excessive agency, DoS, supply chain.
14
The Generative AI Application Lifecycle
40 min · LLMOps extends MLOps for apps built on pretrained models.

Beyond prompts: RAG, models, agents and fine-tuning

15
Retrieval-Augmented Generation and Vector Databases
55 min · RAG retrieves relevant content and adds it to the prompt to ground answers.
16
Open Source Models
40 min · Open models publish weights; licences vary and must be checked.
17
AI Agents
40 min · Agents plan steps, use tools, observe results and iterate toward a goal.
18
Fine-Tuning Your LLM
40 min · Fine-tuning further trains a pretrained model on task examples.
19
Small Language Models
35 min · SLMs have fewer parameters but can do well on focused tasks.
20
Building with Mistral Models
35 min · Mistral AI offers models via its platform and catalogues like Azure AI Foundry and GitHub Models.
21
Building with Meta Models
35 min · Llama is Meta's widely used family of open-weight models.

Course content is adapted from Microsoft's "Generative AI for Beginners" curriculum (https://github.com/microsoft/generative-ai-for-beginners), used under the MIT License, Copyright (c) Microsoft Corporation. Study guides and quiz questions are written by Vikosha. Certificates are issued by Vikosha and are not issued or endorsed by Microsoft. Licence: MIT License, Copyright (c) Microsoft Corporation. View the original curriculum