# AI Fundamentals 101 **A plain-language guide to core AI concepts — from machine learning to agents, written from a software engineer's perspective.** ## Why I Wrote This Series When I started building AI-powered systems, I ran into a problem that no tutorial warned me about: **I didn't actually understand the fundamentals.** I could call an OpenAI API and get a response. I could copy-paste a LangChain example and make a chatbot. But when something went wrong — when my RAG pipeline returned irrelevant results, when my agent got stuck in a loop, when my model's accuracy tanked on new data — I didn't have the foundational knowledge to diagnose the problem. The issue was that most "AI fundamentals" content falls into two camps: either it's academic papers full of math notation, or it's marketing fluff that tells you AI will change the world without explaining *how* it actually works. There was nothing in between for engineers who need to understand the concepts well enough to build real systems. So I wrote this series. Every concept is explained through the lens of "why does this matter when you're building something?" I use Python examples to make abstract ideas concrete, and I reference my own projects — home lab monitoring, personal knowledge bases, DevOps automation — instead of made-up business scenarios. This series is your map of the AI landscape before you start building. *** ## How This Fits with Other Series | Series | Focus | | --------------------------------------------------------------------------------------------------------------------------------- | ----------------------------------------------------------------- | | **AI Fundamentals 101 (this)** | The concepts — what everything is and how it fits together | | [Machine Learning 101](https://blog.htunnthuthu.com/ai-and-machine-learning/machine-learning-101) | Hands-on ML with scikit-learn — algorithms, evaluation, pipelines | | [AI Engineer 101](https://blog.htunnthuthu.com/ai-and-machine-learning/artificial-intelligence/ai-engineer-101) | The AI engineer role — tooling, LLMs, embeddings, production APIs | | [PyTorch 101](https://blog.htunnthuthu.com/ai-and-machine-learning/artificial-intelligence/pytorch-101) | Deep learning from scratch — tensors, autograd, neural networks | | [LLM API Development 101](https://blog.htunnthuthu.com/ai-and-machine-learning/artificial-intelligence/llm-api-development-101) | Production LLM applications with Claude and FastAPI | | [RAG 101](https://blog.htunnthuthu.com/ai-and-machine-learning/artificial-intelligence/rag-101) | End-to-end retrieval-augmented generation with pgvector | | [AI Agent Development 101](https://blog.htunnthuthu.com/ai-and-machine-learning/artificial-intelligence/ai-agent-development-101) | Building agents with ReAct loops, memory, and tool use | **Read this series first** if you're new to AI. It gives you the vocabulary and mental models that every other series assumes you have. *** ## What You Will Learn ### Part 1: What is Artificial Intelligence? * The definition of AI — and why most definitions are wrong * A brief history: symbolic AI → machine learning → deep learning → generative AI → agentic AI * The 7 types of AI: reactive, limited memory, theory of mind, self-aware, ANI, AGI, ASI * Key terminology every engineer needs: models, training, inference, parameters, weights ### Part 2: Machine Learning, Deep Learning, and Foundation Models * Machine learning: learning from data instead of writing rules * Supervised, unsupervised, and reinforcement learning — with Python examples * Deep learning: neural networks and what makes them "deep" * Foundation models: pre-trained, general-purpose models that changed everything * Ten real-world ML use cases you interact with daily ### Part 3: Natural Language Processing — NLP, NLU, and NLG * What is NLP and why it's the backbone of modern AI * NLP vs NLU vs NLG — the processing pipeline explained * Tokenization, stemming, and text preprocessing with Python * Named entity recognition, sentiment analysis, and text classification * From rule-based chatbots to LLM-powered assistants ### Part 4: Large Language Models and Generative AI * What makes a language model "large" — parameters, data, and compute * How transformers work — attention is all you need, in plain language * Generative AI: text, image, code, and multimodal generation * The cost equation: why LLMs are expensive and what prompt caching solves * Limitations: hallucination, reasoning gaps, and the knowledge cutoff problem ### Part 5: RAG, Fine-Tuning, and Prompt Engineering * Three strategies for customizing AI: RAG, fine-tuning, and prompt engineering * When to use each and the trade-offs * RAG explained: retrieval + generation for grounded answers * Multimodal RAG: going beyond text * Practical comparison with Python examples ### Part 6: AI Agents and Communication Protocols * What makes an AI agent different from a plain LLM call * Agent architectures: ReAct, tool use, and planning loops * MCP vs API vs gRPC — how agents connect to tools and data * A2A vs MCP — agent-to-agent vs agent-to-tool communication * Human-in-the-loop: when AI should ask before acting ### Part 7: The AI Stack and Building Real AI Systems * The modern AI stack: hardware → models → frameworks → applications * Adding AI to existing applications — embedded AI patterns * Why most AI projects fail (and how to avoid the "AI graveyard") * NeuroSymbolic AI: combining neural networks with logical reasoning * Responsible AI: bias, fairness, transparency, and accountability *** ## Prerequisites * Basic programming knowledge (Python preferred) * Curiosity about how AI works under the hood * No math background required — every concept is explained in plain language If you've read [Python 101](https://blog.htunnthuthu.com/getting-started/programming/python-101), you're ready. *** ## Stack | Tool | Purpose | | -------------------- | --------------------------- | | Python 3.12 | All code examples | | scikit-learn | ML demonstrations | | NLTK / spaCy | NLP examples | | transformers | Hugging Face model examples | | matplotlib / seaborn | Visualizations | ``` # requirements.txt scikit-learn>=1.5 nltk>=3.9 matplotlib>=3.9 seaborn>=0.13 numpy>=2.0 pandas>=2.2 ``` *** ## Series Structure | Part | Title | Key Topics | | --------------------------------------------------------------------------------------------------------------------------------------------------- | ------------------------------------------------------ | --------------------------------------------- | | [Part 1](https://blog.htunnthuthu.com/ai-and-machine-learning/artificial-intelligence/ai-fundamentals-101/part-1-what-is-artificial-intelligence) | What is Artificial Intelligence? | History, types, terminology, landscape | | [Part 2](https://blog.htunnthuthu.com/ai-and-machine-learning/artificial-intelligence/ai-fundamentals-101/part-2-ml-dl-foundation-models) | Machine Learning, Deep Learning, and Foundation Models | ML types, neural networks, pre-trained models | | [Part 3](https://blog.htunnthuthu.com/ai-and-machine-learning/artificial-intelligence/ai-fundamentals-101/part-3-nlp-nlu-nlg) | Natural Language Processing — NLP, NLU, and NLG | Text processing, NER, sentiment, chatbots | | [Part 4](https://blog.htunnthuthu.com/ai-and-machine-learning/artificial-intelligence/ai-fundamentals-101/part-4-llms-and-generative-ai) | Large Language Models and Generative AI | Transformers, tokens, Gen AI, cost, limits | | [Part 5](https://blog.htunnthuthu.com/ai-and-machine-learning/artificial-intelligence/ai-fundamentals-101/part-5-rag-finetuning-prompt-engineering) | RAG, Fine-Tuning, and Prompt Engineering | Three customization strategies compared | | [Part 6](https://blog.htunnthuthu.com/ai-and-machine-learning/artificial-intelligence/ai-fundamentals-101/part-6-ai-agents-and-protocols) | AI Agents and Communication Protocols | Agents, MCP, A2A, gRPC, human-in-the-loop | | [Part 7](https://blog.htunnthuthu.com/ai-and-machine-learning/artificial-intelligence/ai-fundamentals-101/part-7-ai-stack-and-building-systems) | The AI Stack and Building Real AI Systems | AI stack, embedded AI, failure modes, ethics | *** ## Reference This series uses IBM Technology's [AI Fundamentals playlist](https://www.youtube.com/playlist?list=PLOspHqNVtKADfxkuDuHduUkDExBpEt3DF) as a reference for topic coverage and structure, combined with hands-on experience from personal projects. *** *Let's start with the basics:* [*Part 1 — What is Artificial Intelligence?*](https://blog.htunnthuthu.com/ai-and-machine-learning/artificial-intelligence/ai-fundamentals-101/part-1-what-is-artificial-intelligence)