When I rebranded the blog, you might know that I said "So, what I wanted to tell is that just because I "UPDATED" the blog, it doesn't mean that I'm never gonna post about tech. I will post about it, but I will also post about the other things that I feel passionate about".

So I decided to actually let this post be about one of my passions, about tech. And you might remember that when I started this blog a year ago, I did talk about AI, but only its impact, not how to learn it, work with it, or Machine Learning. So I decided to post about AI&ML.

Artificial Intelligence is everywhere these days.

When we hear “AI,” we might immediately think of ChatGPT, robots, self-driving cars, image generators, or futuristic movies. But AI is actually much bigger than that. It is already part of many of the apps and technologies we use every single day. ✨

I wanted to create a little introduction to AI and Machine Learning that makes these topics easier to understand; without making them feel overly complicated.

So, let’s get into it. 💻🤍

🧠 Introduction to Artificial Intelligence and Machine Learning

🌎 Discovering AI in Our World

AI might sound like something from the future, but it is already around us.

From recommendations on YouTube and Netflix to voice assistants, search engines, face recognition, translation tools, and social media algorithms, AI is quietly working behind the scenes.

The interesting part is that we don't always notice it.

💭 What Does “Intelligence” Mean?

Before understanding artificial intelligence, it helps to think about intelligence itself.

Humans can learn, solve problems, recognize patterns, understand language, make decisions, and adapt to new situations.

AI tries to recreate some of these abilities using computers and algorithms.

🤖 What Is AI? A Quick Trip Through History

Artificial Intelligence is the field of creating computer systems that can perform tasks that normally require human-like intelligence.

The idea isn't actually new. Researchers have been thinking about machine intelligence for decades, from early computer programs to today's advanced AI systems.

The technology has changed dramatically, but the basic question remains:

Can machines learn, reason, and perform tasks that we normally associate with human intelligence?

⚙️ How Does AI Really Work?

At a basic level, AI systems use algorithms, data, and computing power to recognize patterns and produce results.

For example, if an AI system is trained using thousands of pictures of different animals, it can learn patterns that help it identify new pictures.

And that brings us to Machine Learning. 👀

🎯 Activity: Is It AI or Not?

Here's a fun challenge:

A calculator adding two numbers
A phone recognizing your face
A recommendation system suggesting a video
A basic alarm clock
A chatbot answering questions

Which ones would you consider AI?

Sometimes the answer isn't as obvious as it seems!

🧩 Types of AI

AI can be discussed in different ways, but one common distinction is between narrow AI, which is designed for specific tasks, and the idea of general AI, which would be capable of handling a much wider range of intellectual tasks.

Most AI systems we use today are specialized systems.

🔍 Subfields of AI

AI isn't just one technology. It includes many different areas, including:

Machine Learning
Natural Language Processing
Computer Vision
Robotics
Knowledge Representation
Generative AI

Each area focuses on a different way computers can interact with information and the world.

📚 Introduction to Machine Learning

🤔 What Is Machine Learning?

Machine Learning is a major part of AI.

Instead of programming a computer with every single rule for a task, Machine Learning allows a system to learn patterns from data.

For example, instead of telling a computer exactly what every picture of a cat looks like, we can give it many examples and allow a model to learn useful patterns from them.

⚙️ How Does Machine Learning Work?

A simplified Machine Learning process looks something like this:

Data → Training → Model → Prediction

The model looks for patterns in training data and then uses what it learned to make predictions or decisions about new data.

🧩 Types of Machine Learning

Some of the main types are:

Supervised Learning — the model learns from labeled examples.

Unsupervised Learning — the model looks for patterns or structures in data without predefined labels.

Reinforcement Learning — a system learns through interaction, using rewards and feedback.

☁️ Using AI: Local vs. Cloud-Based

So, where does AI actually run?

Not every AI system works in exactly the same way.

☁️ Cloud-Based AI

With cloud-based AI, the processing happens on remote servers.

You send information to a service, the service processes it, and the result comes back to you.

This is one reason APIs are so important in modern AI.

🌐 Exploring Google Cloud AI

Cloud platforms provide access to AI tools without requiring everyone to own powerful computing hardware.

This makes it possible for developers and organizations to experiment with machine learning and AI services more easily.

💻 Local AI

Local AI means running an AI model directly on your own computer or device.

This can provide more control over your environment and can be useful when you want processing to happen locally.

🤗 Exploring Hugging Face

Hugging Face is a major part of the modern open AI ecosystem, with models, datasets, and tools that developers and learners can explore.

📱 Edge AI

Edge AI takes the idea of local processing even further by running AI directly on devices such as phones, cameras, sensors, and other hardware.

⚖️ Cloud vs. Local vs. Edge

Each approach has different advantages.

Cloud: powerful and scalable ☁️
Local: more control and local processing 💻
Edge: AI closer to where the data is created 📱

🔌 Introduction to APIs

🤔 What Is an API?

API stands for Application Programming Interface.

Basically, an API allows different pieces of software to communicate with each other.

Think of it like a waiter in a restaurant:

You make a request → the waiter takes it to the kitchen → the kitchen processes it → the waiter brings the result back.

⚙️ How Do APIs Work?

An application sends a request to an API.

The API communicates with another service and returns a response.

This is one of the main ways developers can add AI capabilities to their own applications.

🤖 Why Does Big AI Hide Behind APIs?

Modern AI models can be extremely large and computationally expensive.

Instead of running everything yourself, an API can allow your application to communicate with a powerful AI service.

💻 Using an API in Your Project

Once you understand the basic idea of APIs, you can start connecting AI capabilities to your own applications.

And that's where things get really interesting. 👀

✨ Generative Artificial Intelligence

Generative AI is one of the most exciting areas of AI right now.

Instead of simply analyzing existing information, generative models can create new content such as:

Text
Images
Audio
Video
Code

🧠 Introduction to Neural Networks

Neural networks are computing systems inspired loosely by the way biological neural networks process information.

They contain layers of interconnected units that transform information and learn patterns from data.

⚙️ How Does Generative AI Work?

Generative AI models learn patterns from large amounts of training data.

Once trained, they can use those learned patterns to generate new outputs based on an input or prompt.

🛠️ Techniques Used in Generative AI

Different architectures and techniques are used for different types of generation.

Some examples include transformer-based models, diffusion models, and other neural network architectures.

And yes; there's a LOT more happening behind the scenes than simply typing a prompt and pressing enter. 😭💻

⚠️ Responsible Use of AI

AI is powerful, but that doesn't mean everything it produces is automatically correct or harmless.

🤍 Introduction to AI Ethics

AI ethics is about thinking carefully about how AI should be developed and used.

Important questions include:

Is the system fair?
Is people's privacy protected?
Can we understand how decisions are made?
Who is responsible when something goes wrong?

🌱 Principles of Ethical AI

Responsible AI often focuses on ideas such as:

Fairness • Transparency • Privacy • Safety • Accountability

🚨 When AI Goes Wrong

AI systems can make mistakes.

They can produce inaccurate information, reflect biases in their training data, or be used in ways their creators didn't intend.

That's why human judgment still matters.

📱 Responsible Digital Citizenship

Using AI responsibly also means thinking before sharing information, checking important claims, respecting other people's work, and understanding that AI-generated content isn't automatically trustworthy.

💬 Natural Language Processing (NLP)

How can a computer work with human language?

That's where Natural Language Processing, or NLP, comes in.

NLP focuses on helping computers process and work with human language.

📜 From Rules to Neural Networks

Early NLP systems often relied heavily on manually written rules.

Later, statistical Machine Learning approaches became popular.

Then, during the 2010s, deep learning and neural networks transformed modern NLP.

Today, neural networks and transformer-based architectures are central to many language systems.

📝 NLP Tasks

NLP can be used for tasks such as:

Translation
Text classification
Sentiment analysis
Question answering
Summarization
Text generation
Speech and language applications

🧠 Large Language Models (LLMs)

You've probably heard the term LLM everywhere lately.

LLM stands for Large Language Model.

These models are designed to process and generate human language.

📚 How Do LLMs Learn?

During training, an LLM processes huge amounts of text and learns statistical patterns in language.

It doesn't simply memorize a giant encyclopedia. It learns relationships between pieces of information and language patterns that allow it to generate responses.

🔄 Types of LLMs

Different language models can have different architectures, sizes, training methods, capabilities, and purposes.

Some are designed primarily for generating text, while others can work with multiple types of information.

💬 Interacting With an LLM

The way we communicate with an LLM matters.

The instructions we give it are called prompts, which leads us to another important topic:

Prompt Engineering. 👀

⚖️ Strengths and Limits of LLMs

LLMs can be incredibly useful for brainstorming, explaining concepts, summarizing information, writing code, and many other tasks.

But they can also make mistakes.

They don't automatically know whether every statement they generate is true, so important information should always be checked.

✍️ Prompt Engineering

💡 What Is Prompt Engineering?

Prompt engineering is the process of designing instructions that help an AI system produce a useful result.

A good prompt can provide:

Context
Clear instructions
Desired format
Examples
Constraints

🧠 Key Prompt Engineering Techniques

Instead of simply saying:

“Tell me about AI.”

you could give the AI a specific goal, audience, format, and level of detail.

The more clearly you communicate what you want, the easier it is for the model to produce a useful response.

❌ Common Misconceptions

Prompt engineering isn't about finding a magical sentence that makes AI perfect.

It's more about clear communication, experimentation, and iteration.

🌟 Skills of a Good Prompt Engineer

Curiosity, critical thinking, communication, experimentation, and understanding the strengths and limitations of AI are all valuable skills.

💻 Running a Local LLM

Want to experiment with AI on your own computer?

Local LLMs make that possible.

🖥️ Ways to Run LLMs Locally

There are different tools and approaches for running models locally depending on your computer's hardware and the model you want to use.

🦙 Introduction to Ollama

Ollama is one tool that makes experimenting with local language models more accessible.

You can use it to download and run supported models locally and interact with them from your computer.

🤖 Running a Model With Ollama

Once everything is set up, you can interact with a local model and start experimenting with prompts; without needing to send every interaction to a cloud-based service.

It's a pretty cool way to understand what's happening behind the scenes. 👀

💬 Building Your Own AI Chatbot

Okay, now we're getting into the fun part.

Instead of only using AI tools, we can start building something ourselves.

Using Python and a local model, we can create a simple chatbot application.

We can experiment with:

Chatbot personalities
System instructions
Prompt engineering
Creativity
Response style
Model parameters

Changing different settings can make an AI response feel more focused, creative, formal, or conversational.

Basically, you're not just talking to AI; you're learning how to shape the experience. 🤖✨

👁️ Computer Vision: Perception and Generation

AI isn't limited to text.

Computers can also work with images and video.

📸 What Is Computer Vision?

Computer Vision is an area of AI focused on helping computers understand visual information.

👀 Perception vs. Generation

Perception means analyzing or understanding visual information.

Generation means creating new visual content.

🔍 Perception Tasks

Computer Vision can be used for tasks such as:

Image classification
Object detection
Face detection
Image segmentation
Optical character recognition

🌎 Real-World Applications

Computer Vision is used in areas such as healthcare, transportation, security, manufacturing, accessibility, and many everyday technologies.

🎨 Image Generation

Generative AI can also create images from text descriptions.

Instead of analyzing an existing picture, the model generates a new visual output based on the instructions it receives.

🖼️ Image Classification

Image classification sounds complicated, but the basic idea is pretty simple:

Give an image to a model → the model predicts what category it belongs to.

🔄 The Basic Process

A typical image classification workflow includes:

Dataset → Training → Model → Testing → Prediction

🧠 Pre-Trained Models

Instead of training a model completely from scratch, developers can sometimes use pre-trained models and adapt them for specific tasks.

🧪 Teachable Machine

Tools such as Teachable Machine make it easier to experiment with image classification without needing to build everything from scratch.

It's a great way to see Machine Learning in action.

♻️ Building a Waste Classification App

One practical project is creating an app that can recognize different categories of waste.

📊 Part 1: Dataset & Model Training

First, we need data.

We can collect or find suitable images, organize them into categories, and use them to train an image classification model.

The quality and variety of the dataset matter a LOT because the model can only learn from the examples it receives.

🖥️ Part 2: Building the App UI

After training the model, we can connect it to a Python application.

Using Tkinter, we can create a simple interface where users can provide an image and receive a classification result.

Then comes the important part:

Testing.

We need to see how well the model performs on images it hasn't seen before.

🔊 Introduction to Text-to-Speech

What if AI could not only understand text but also speak it?

That's where Text-to-Speech (TTS) comes in.

🗣️ What Is TTS AI?

Text-to-Speech technology converts written text into spoken audio.

⚙️ How Does TTS Work?

A TTS system processes written language and generates speech that represents the text.

Modern systems can produce increasingly natural-sounding voices.

🤔 Why Do We Use TTS?

TTS can be useful for:

Accessibility
Education
Virtual assistants
Apps
Navigation
Entertainment
Reading digital content aloud

🔍 Google TTS and Mozilla TTS

There are different TTS technologies and systems available, each with different approaches, voices, and capabilities.

Exploring real systems is a great way to understand how AI-powered speech works outside of theory.

🎙️ Integrating TTS Into Applications

The next step is actually putting TTS into a project.

We can connect text-to-speech technology to an application and make the program speak its responses.

Tools such as Google Text-to-Speech can be used for this kind of project, while other platforms and open-source tools can provide additional possibilities.

And that's basically the bigger idea behind this whole AI & ML journey:

Don't just use the technology. Learn how it works. Experiment with it. Build something with it. 🤍💻

🌱 Final Thoughts

AI and Machine Learning can seem intimidating at first because there are so many new words, technologies, and concepts.

But you don't need to understand everything at once.

Start with the basics.

Learn what AI is.

Understand how Machine Learning works.

Experiment with simple models.

Try an API.

Build a chatbot.

Play around with Computer Vision.

Explore text-to-speech.

And most importantly; stay curious.

The AI world is changing incredibly quickly, and there is always something new to learn.

Maybe the coolest part isn't just seeing what AI can do.

Maybe it's realizing that you can learn to build with it too. ✨🤖💻.

Soon, I'm gonna be posting a tutorial step-by-step on how to work with AI; not just theory, but also working and discovering too. For now, I have nothing left to say so until next time, byee. 🤍