First Step into AI Mastery: LLMs & Prompt Engineering

Prompt Engineering Crash Course

Created byAvesta Jafari

Last updated 4/2026

English

What you'll learn

Understand how Large Language Models (LLMs) work, including next-token prediction, embeddings, and similarity-based reasoning
Master prompt engineering techniques (zero-shot, few-shot, chain-of-thought) to control and improve AI outputs
Run and use local LLMs with Ollama, including selecting the right models and optimizing performance
Optimize LLM applications for performance, accuracy, and scalability in production

Course content

6 sections • 14 lectures • 1h 47m total length

Course Overview, Roadmap & Instructor Introduction1:50
Course Overview, Roadmap & Instructor Introduction
In this first video, you’ll get a clear overview of what this course is all about and how it will help you become a GenAI engineer.
I’ll introduce myself, share my background, and explain why this course is designed the way it is — focusing on real-world skills, not just theory.

Installing Envireoment by Code and 2 Technique of Propmt Engineerig17:51
1. Zero-Shot Prompt
Description:
Give the model a task directly without examples. It relies only on general knowledge.
2. Few-Shot Prompt
Description:
Provide a few examples before the task so the model learns the pattern.
How we talk with models! 3rd and 4th Example of Prompt Engineering3:42
3. Instruction Prompt
Description:
Clearly tell the model what to do using direct commands or rules.
4. Role-Based Prompt
Description:
Assign a role (e.g., teacher, doctor, developer) to shape tone and expertise.
5th, 6th ,7th ,8th16:58
5. Chain-of-Thought Prompt
Description:
Ask the model to explain step-by-step reasoning before giving the answer.
6. Output-Format Prompt
Description:
Specify the exact format (JSON, table, list, etc.) for structured output.
7. Contextual Prompt
Description:
Provide background or context so the model gives more accurate answers.
8. Constraint-Based Prompt
Description:
Limit the response using rules (length, tone, style, or restrictions).
9th Navigate prompting6:59
Navigate Prompting = knowing how to “steer” AI to get exactly what you want.
Knowledge prompting and RAG Style Prompting14:23
Knowledge Prompting VS RAG Style Prompting:
Knowledge Prompting is a technique where you guide the AI by injecting relevant knowledge directly into the prompt.

RAG Style Prompting goes one step further.

It combines:

Retrieval → Fetching relevant data from external sources (database, documents, APIs)
Generation → Using that data to generate a response

The AI doesn’t rely only on memory—it uses real, up-to-date, external knowledge.
Automatic Prompting and Also Prompt Chaining12:08
In this video, we explore Automatic Prompt Engineering and Prompt Chaining using modern LLM workflows.
You will learn how to transform simple prompts into more precise and effective ones, and how to build multi-step pipelines that produce structured and useful outputs.
What you will learn:
How to improve prompts automatically using a meta-prompt
Converting vague questions into clear and optimized prompts
Building a prompt chain step by step
Summarizing text using LLMs
Generating actionable steps from summaries
Combining multiple chains into a single workflow
Example covered:
We start with a simple question like "why is the sky blue" and improve it using an LLM. Then we create a pipeline that first summarizes input text and then generates actionable steps based on that summary.
This video is suitable for developers, AI engineers, and anyone interested in prompt engineering and building LLM-based applications.
Keywords:
AI, LangChain, Prompt Engineering, Prompt Chaining, LLM, Python, Artificial Intelligence
Skeleton-of-Thought (SoT) and Tree of Thoughts (ToT)12:07

Requirements

Basic familiarity with Python or any programming language is helpful, required.
A computer capable of running local AI models (LLMs) (atleast 4GB ram).

Description

n this course, you’ll get a complete, structured roadmap to mastering modern AI development from the ground up. We’ll start by breaking down how Large Language Models (LLMs) actually work, so you move beyond using them as black boxes and truly understand what’s happening under the hood. From there, you’ll learn how to master prompt engineering—one of the most critical skills for getting accurate, reliable, and high-quality outputs from AI systems.

Next, we’ll go hands-on with running models locally using tools like Ollama, giving you more control, privacy, and flexibility. You’ll then learn how to build powerful Retrieval-Augmented Generation (RAG) systems using your own data, enabling real-world, customized AI applications.

We’ll also dive into designing and implementing AI agents that can reason, plan, and take actions autonomously. Finally, we’ll guide you step-by-step on how to move from simple demos to production-ready applications that are scalable and robust.

You’ll also see the full roadmap of the course, including:

How LLMs (Large Language Models) actually work
How to master prompt engineering
How to run models locally using Ollama
How to build RAG systems with your own data (introduction)
How to design and implement AI agents
How to move from simple examples to production-ready applications

By the end of this video, you’ll understand exactly what you’re going to build and how each part of the course connects together.

Let’s get started :

Copun code: 80% 453536ED50DA0F6A487A

Who this course is for:

For People who with low resources looking for free, effective AI tools

First Step into AI Mastery: LLMs & Prompt Engineering

What you'll learn

Explore related topics

Course content

Introduction1 lecture • 2min

LLMs1 lecture • 6min

Ollama3 lectures • 10min

VSCode1 lecture • 4min

Propmpt Engineering7 lectures • 1hr 24min

Real-World Project !1 lecture • 1min

Requirements

Description

Who this course is for: