AI For Beginners - 30 Days From Zero to Your First AI App

AI For Beginners - 30 Days From Zero to Your First AI App

By @subraatakumar | Learn AI in Simple English, With Chai and Solan Stories

---

1. What is This Book?

This is not another boring textbook with heavy math.

AI For Beginners is a 30-day, hands-on book that takes you from "What is AI, really?" to "I built my own AI app that answers Solan tourism questions."

Every day, 1 chapter. Every chapter: a small story from Solan, a simple definition with no jargon, how it actually works, a real code lab where you train a model and save it as model.pkl, and a mini project you can finish before your chai cools.

By Day 30, you won't just know AI, you will have built it.

Core Idea of the Book: Every AI follows the same 3 simple steps — Data, Model, Training. We repeat this pattern for 30 days until it becomes muscle memory.

---

2. Who Is This For?

This book is for you if:

• You have never built AI before, you are an absolute beginner
• You know basic Python (if, for loop, pip install) — that's enough
• You use ChatGPT but don't understand why it sometimes makes up stuff
• You are a student in Solan, Shimla, Delhi, or anywhere, and want to understand AI without heavy math
• You are a creator, shop owner, teacher, farmer, developer who wants to use AI in daily work
• You tried other AI courses and got lost in jargon like "transformers", "embeddings" on Day 1

This book is NOT for:

• PhD researchers looking for math proofs (we explain intuition, not equations)
• People who want to train a ChatGPT from scratch with billions of dollars (we use existing LLMs + RAG)
• People who want only theory with no code (every chapter has train.py + predict.py)

Prerequisites:

• Python installed, can run python3 --version
• Can run pip install
• Curiosity and 30 minutes a day

---

3. How This Book Is Structured

30 Chapters, 9 Clean Sections Per Chapter — Same Pattern Every Day (Option A Style):

Every chapter follows this 9-section template, so your brain knows what to expect:

1. A Morning in Solan — Hook story from Solan bus stand, Mall Road, Karol Tibba, chai tapri. Real life, not theory.
2. What X Really Is — Simple definition in 2-3 lines, no jargon, promise.
3. How It Works — The actual mechanism, in points, with examples.
4. The Analogy — Child and Dog, Kitchen, Chai Tapri, Photocopy — makes it stick.
5. Code Lab: Train and Predict — Real code, 2 files: train.py (creates + saves model.pkl) + predict.py (loads + predicts). Train once, use 1000 times — like real companies.
6. AI Around You — Where you already see this in Instagram, YouTube, UPI, Google Maps.
7. The Common Mistake — What 90% beginners do wrong.
8. Mini Project — 5-10 minute task you can do today.
9. Recap and Challenge — One line summary + challenge to test your friend.

Plus for every chapter:

• 3 Premium Images (1080x1080, dark gradient, glassmorphism, watermark @subraatakumar) — placed after relevant sections, not all at top.
• Codelab Folder — codelab/chapter-XX-name/starter/ with train.py, predict.py, requirements.txt + solution/ folder.

---

4. The 30 Chapters — What You Will Learn

We go from zero to your first AI app. Each chapter builds on previous.

Phase 1: Foundations (Day 1-6) — What is AI?

Phase 2: How LLMs Actually Work (Day 7-13)

Phase 3: Advanced Building Blocks (Day 14-22)

Phase 4: Real-World AI Engineering (Day 23-30)

---

5. How To Use This Book + Codelab

One .venv for Whole Book (Not 30):

cd codelab
python3 -m venv .venv
source .venv/bin/activate
# You will see (.venv) in terminal

For Each Chapter:

pip install -r chapter-01-what-is-ai/starter/requirements.txt
python chapter-01-what-is-ai/starter/train.py   # creates model.pkl
python chapter-01-what-is-ai/starter/predict.py  # uses model.pkl

Train Once, Predict 1000 Times Pattern:

• train.py = Heavy, learns, saves model.pkl file to disk
• predict.py = Light, loads model.pkl, predicts fast, no retraining

This is how real companies do it. You learn correct architecture from Day 1.

Requirements: Each chapter has own requirements.txt + global requirements-all.txt. We use >= not == so book stays future-proof.

---

6. What Makes This Book Different?

1. Solan Stories, Not Silicon Valley: Every chapter starts at Solan bus stand, Mall Road, Karol Tibba, chai tapri — not San Francisco.
2. No Jargon Without Explanation: If we say "embedding", we explain next line: "Meaning to numbers".
3. 2-File CodeLab: Not 1 messy main.py. Train and predict separated, model.pkl persistence explained — answers "Where does model file stay?"
4. 3 Premium Images Per Chapter: Not flat white placeholders. Dark gradient, glassmorphism, 1080x1080, like Instagram carousels you liked.
5. Option A Clean Headings: No awkward "Hook Story (A Solan Morning)" — just "A Morning in Solan". Book-like, not blog-like.
6. Best Approach, Not Blind: If something is not best practice, we argue and fix it, not just follow.

---

7. Daily Instagram Carousels

This book is paired with Daily AI Carousels for Beginners (5 slides per day, same content as book but carousel format).

Check /Daily_AI_carousels_for_beginners.md for Day 2-4 examples. Same style, same watermark @subraatakumar, same Solan tone.

Book = Deep dive, Carousels = Quick recap for Instagram.

---

8. About The Author

@subraatakumar — Subraata Kumar

Builder, teacher, storyteller from Solan, Himachal Pradesh. Believes AI should be explained like you explain to chai wala at bus stand — simple enough that he gets it before chai cools.

Find more: Instagram @subraatakumar, @dashbabupresents

Location: Solan, Himachal Pradesh, India — all stories from here.

---

9. Final Note — Before You Start

You don't need to be smart to learn AI. You need to be consistent.

30 days, 30 minutes a day, 1 chapter a day. By Day 30, you will have your own Solan Tourism AI Assistant that you built — not copied.

Open Chapter 1: What is AI, Really? — Let's start with that foggy morning in Solan.

Every AI follows the same 3 simple steps. Let's learn them.

---