10 Days · 2 Hours / Day · Full Notes

10-Day AI Internship — Full Notes

Complete day-by-day study notes for Artificial Intelligence — from Python basics to building your own AI model and capstone project in 10 days.

Certificate IncludedTheory + Practical DailyOnline & Offline

Jump to a day

Day 1: AI FoundationsDay 2: Python Essentials for AIDay 3: Working with Data — NumPy & PandasDay 4: Data VisualizationDay 5: Machine Learning BasicsDay 6: Build Your First ModelDay 7: Model Evaluation & ImprovementDay 8: Introduction to Neural NetworksDay 9: Generative AI & AI ToolsDay 10: Capstone Project & Certification
1

Day 1

AI Foundations

Understand what Artificial Intelligence really is and how it powers the apps you use every day.

Theory — Hour 1

What is Artificial Intelligence?

  • AI is the science of making computers perform tasks that normally need human intelligence — like understanding language, recognising images, or making decisions.
  • Narrow AI: built for one task (e.g. spam filters, face unlock). This is all AI we use today.
  • General AI: a hypothetical machine that can do any intellectual task a human can. It does not exist yet.

AI vs Machine Learning vs Deep Learning

  • AI is the broadest idea — any machine that mimics human intelligence.
  • Machine Learning (ML) is a part of AI where machines learn patterns from data instead of being explicitly programmed.
  • Deep Learning (DL) is a part of ML that uses neural networks with many layers to learn very complex patterns (images, speech, text).

Types of Machine Learning

  • Supervised learning: learn from labelled data (e.g. emails marked spam / not spam).
  • Unsupervised learning: find hidden groups in unlabelled data (e.g. customer segments).
  • Reinforcement learning: learn by trial and error with rewards (e.g. game-playing bots).

Real-World Applications

  • Healthcare: disease detection from scans. Finance: fraud detection. Retail: product recommendations.
  • Everyday: Google Maps routing, YouTube/Netflix recommendations, voice assistants, ChatGPT.

Practical — Hour 2

  • Explore 3 popular AI tools (ChatGPT, an image generator, Google Lens) and note what each one does.
  • List 5 apps on your phone that use AI and identify which task the AI performs.

Key Takeaways

  • AI > Machine Learning > Deep Learning (each is a subset of the one before).
  • All AI today is 'narrow' — built for specific tasks.
  • ML learns from data; the more good data, the better it performs.
2

Day 2

Python Essentials for AI

Learn the core Python you actually need to start building AI — nothing extra.

Theory — Hour 1

Why Python for AI?

  • Simple, readable syntax — beginners can focus on logic, not complexity.
  • Huge ecosystem of AI libraries (NumPy, Pandas, scikit-learn, TensorFlow).
  • The default language for almost all AI/ML work in industry and research.

Variables & Data Types

  • int (numbers), float (decimals), str (text), bool (True/False).
  • Variables store values: name = 'Sudhar', age = 20.

Lists & Dictionaries

  • List: an ordered collection — marks = [85, 90, 78].
  • Dictionary: key-value pairs — student = {'name': 'Priya', 'dept': 'CSE'}.

Loops, Conditions & Functions

  • if/else makes decisions; for/while repeat actions.
  • Functions group reusable code: def greet(name): return 'Hi ' + name.

Your first AI-style program

marks = [85, 90, 78, 92, 70]
average = sum(marks) / len(marks)

if average >= 80:
    print('Result: Distinction')
else:
    print('Result: Pass')

Practical — Hour 2

  • Write a program that stores 5 students' marks and prints the highest, lowest and average.
  • Create a dictionary for yourself (name, department, year) and print each value.

Key Takeaways

  • Python is the #1 language for AI because it is simple and library-rich.
  • Master variables, lists, dictionaries, loops and functions — that covers 90% of AI coding basics.
3

Day 3

Working with Data — NumPy & Pandas

Data is the fuel of AI. Learn to load, explore and clean real datasets.

Theory — Hour 1

NumPy — Numerical Python

  • Provides fast 'arrays' for doing maths on lots of numbers at once.
  • Used under the hood by almost every AI library.

Pandas — DataFrames

  • A DataFrame is like an Excel sheet inside Python — rows and columns.
  • Read data with pd.read_csv('data.csv'); view with df.head().

Data Cleaning

  • Real data is messy: missing values, duplicates, wrong formats.
  • Handle missing values with df.dropna() or df.fillna(0).
  • Clean data = better AI models. 'Garbage in, garbage out.'

Load and explore a dataset

import pandas as pd

df = pd.read_csv('students.csv')
print(df.head())          # first 5 rows
print(df.describe())      # quick statistics
df = df.dropna()          # remove missing values

Practical — Hour 2

  • Load a sample CSV dataset, print the first 5 rows and basic statistics.
  • Find and remove any missing values, then count how many rows remain.

Key Takeaways

  • Pandas DataFrames are the standard way to handle data in AI.
  • Cleaning data is often 70% of a real AI project — never skip it.
4

Day 4

Data Visualization

Turn raw numbers into clear visual insights using charts.

Theory — Hour 1

Why Visualize Data?

  • Humans understand pictures faster than tables of numbers.
  • Charts reveal trends, patterns and outliers before you build any model.

Common Chart Types

  • Bar chart: compare categories. Line chart: show change over time.
  • Scatter plot: relationship between two variables. Histogram: distribution of one variable.

Tools

  • Matplotlib: the core plotting library.
  • Seaborn: built on Matplotlib for beautiful charts in fewer lines.

Plot a simple chart

import matplotlib.pyplot as plt

subjects = ['Maths', 'AI', 'Python']
marks = [85, 92, 78]

plt.bar(subjects, marks)
plt.title('My Marks')
plt.show()

Practical — Hour 2

  • Create a bar chart and a line chart from a dataset of your choice.
  • Use a scatter plot to check if two columns (e.g. study hours vs marks) are related.

Key Takeaways

  • Always visualize data before modelling — it saves time and reveals problems.
  • Pick the right chart for the question you are asking.
5

Day 5

Machine Learning Basics

Understand how a machine actually 'learns' from data.

Theory — Hour 1

Features and Labels

  • Features (X): the inputs — e.g. study hours, attendance.
  • Label (y): the answer you want to predict — e.g. pass/fail.

Supervised vs Unsupervised

  • Supervised: you have labelled examples and learn to predict the label.
  • Unsupervised: no labels — the model finds groups/patterns on its own.

The ML Workflow

  • 1. Collect data 2. Clean data 3. Split into train/test
  • 4. Train a model 5. Test it 6. Improve and use it.

Practical — Hour 2

  • For a sample problem (predicting student result), list the features and the label.
  • Draw the 6-step ML workflow and explain each step in your own words.

Key Takeaways

  • ML = learning a mapping from features (X) to a label (y).
  • The workflow is always the same: data → clean → split → train → test → improve.
6

Day 6

Build Your First Model

Train a real prediction model using scikit-learn.

Theory — Hour 1

Regression vs Classification

  • Regression: predict a number (e.g. house price, marks).
  • Classification: predict a category (e.g. pass/fail, spam/not spam).

scikit-learn

  • The most popular beginner-friendly ML library in Python.
  • Every model follows the same pattern: fit() to train, predict() to use.

Train a simple classifier

from sklearn.linear_model import LogisticRegression
from sklearn.model_selection import train_test_split

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2)
model = LogisticRegression()
model.fit(X_train, y_train)        # train
predictions = model.predict(X_test) # use

Practical — Hour 2

  • Train a classification model to predict pass/fail from study hours & attendance.
  • Make predictions on new data and check if they make sense.

Key Takeaways

  • Regression predicts numbers; classification predicts categories.
  • In scikit-learn, every model uses .fit() to learn and .predict() to use.
7

Day 7

Model Evaluation & Improvement

Measure how good your model is and make it better.

Theory — Hour 1

Train/Test Split

  • Train on one part of the data, test on a part the model has never seen.
  • This tells you how it will perform on real, unseen data.

Key Metrics

  • Accuracy: % of correct predictions.
  • Precision & Recall: important when classes are imbalanced.

Overfitting vs Underfitting

  • Overfitting: memorises training data, fails on new data.
  • Underfitting: too simple, performs poorly even on training data.
  • Goal: a balanced model that generalises well.

Measure accuracy

from sklearn.metrics import accuracy_score

preds = model.predict(X_test)
print('Accuracy:', accuracy_score(y_test, preds))

Practical — Hour 2

  • Calculate your model's accuracy on the test set.
  • Try changing the model or features to improve accuracy and record the change.

Key Takeaways

  • Always test on unseen data — accuracy on training data can be misleading.
  • Watch for overfitting; a simpler model that generalises beats a complex one that memorises.
8

Day 8

Introduction to Neural Networks

Discover the technology behind modern deep learning and generative AI.

Theory — Hour 1

Inspired by the Brain

  • Neural networks are loosely modelled on how neurons connect in the brain.
  • Made of layers of connected 'neurons' that pass signals forward.

How It Learns

  • Each connection has a 'weight'. Learning = adjusting weights to reduce error.
  • Activation functions decide whether a neuron 'fires'.

Deep Learning

  • 'Deep' = many layers. Powers image recognition, speech, and ChatGPT.
  • Needs more data and computing power than classic ML.

Practical — Hour 2

  • Use an online neural-network playground to see how layers affect learning.
  • Run a simple neural network on a sample dataset and observe its predictions.

Key Takeaways

  • Neural networks learn by adjusting weights to minimise error.
  • Deep learning = neural networks with many layers; it powers most cutting-edge AI.
9

Day 9

Generative AI & AI Tools

Learn to use ChatGPT and AI tools effectively — a skill every professional needs.

Theory — Hour 1

What is Generative AI?

  • AI that creates new content — text, images, code, audio.
  • Powered by Large Language Models (LLMs) like the ones behind ChatGPT.

Prompt Engineering

  • A 'prompt' is the instruction you give the AI. Better prompts = better results.
  • Be specific: give context, role, format and examples.
  • Bad: 'write about AI'. Good: 'Write a 100-word LinkedIn post about AI for BBA students, friendly tone.'

AI for Productivity

  • Summarise documents, draft emails, generate ideas, write & debug code.
  • Always review AI output — it can be confidently wrong (hallucinations).

Practical — Hour 2

  • Write 3 versions of a prompt for the same task and compare the outputs.
  • Use an AI tool to automate one real task (e.g. summarise notes or draft a report).

Key Takeaways

  • Prompt engineering is the new essential skill — clear, specific prompts win.
  • Always fact-check AI output; treat it as a smart assistant, not a final authority.
10

Day 10

Capstone Project & Certification

Bring everything together by building and presenting your own AI mini-project.

Theory — Hour 1

Plan Your Project

  • Pick a simple, real problem (e.g. predict student results, classify reviews).
  • Decide your data, features, label and the type of model.

Build End-to-End

  • Apply the full workflow: data → clean → visualize → train → evaluate.
  • Document each step so others can understand your work.

Present & Certify

  • Explain your problem, approach, result and what you learned.
  • Receive your QDCODEX AI Internship Completion Certificate.

Practical — Hour 2

  • Complete a mini AI project end-to-end using a dataset of your choice.
  • Prepare a 5-minute presentation: problem, solution, accuracy, learnings.

Key Takeaways

  • A finished project for your portfolio is worth more than any theory.
  • You can now explain the full AI workflow and build a basic model yourself.

What You Will Get

Everything included in the 10-day AI internship

Internship Certificate

Industry-recognised certificate on completion.

Capstone Project

A real AI project for your portfolio.

Hands-On Coding

Practical Python & AI skills, not just theory.

Generative AI Skills

ChatGPT & prompt engineering for real tasks.

Ready to Start Your AI Internship?

Join the 10-day AI internship — online or offline — with a certificate on completion. Open to all departments, no prior experience needed.

See 15-Day ML Plan