AI Engineering

Syllabus AI Academy
 Software
Engineering
Week 1
Python Basics and Data
Structures
1 Introduction, Control 2 Lists,Tuples, 3 List Comprehensions,
Structures, Functions Dictionaries, Sets Generators, Decorators
Introduction to Python and its Lists: creation, indexing, slicing, List comprehensions: syntax and
application methods (append, remove, etc.) use case
Python syntax and indentatio Tuples: creation, indexing, Generators: yield keyword,
Variables and data type immutabilit creating and using generator
Control structures: if-else Dictionaries: key-value pairs, Decorators: function decorators,
statements, loops (for, while) methods (get, keys, values, etc.) applying multiple decorators
Functions: defining functions, Sets: creation, methods (add,
arguments, return values remove, union, intersection, etc.)
Week 2
Python Advanced
Features

1 File I/O, Exception 2 Modules and Packages, 3 Working with APIs,Web

Handlin Standard Library Scraping
File I/O: reading from and writing Modules: creating and importing Working with APIs: making HTTP
to files module requests, handling responses
Exception handling: try, except, Packages: structuring code with Web scraping: BeautifulSoup,
finally blocks, raising exceptions packages, __init__.p requests library
Standard library: overview of
commonly used modules (os,
sys, math, etc.)
Week 3
Python Object-Oriented
Programming (OOP)

1 Introduction 2 Advanced OOP 3 Concurrency (Threading

to OOP Concepts and Multiprocessing)

Introduction to OOP: classes and Advanced OOP concepts: class Introduction to concurrency:

object methods, static methods, threading and multiprocessing

Defining and using classes, properties The threading module: creating

instance methods, and Using decorators with classe and managing thread

attributes Designing with inheritance and The multiprocessing module:

Inheritance and polymorphism composition creating and managing

processes
Week 4
Python Concurrency
(Asynchronous Programming)

1 Introduction to 2 Advanced Asynchronous Integration and

3
Asynchronous Programming Techniques Deployment

Introduction to asynchronous Advanced concurrency techniques: Integrating Python applications

programming: asyncio module futures, coroutines with databases
Defining and running asynchronous task Handling exceptions in asynchronous Deployment strategies:
Using async and await keywords cod packaging and distributing
Using concurrent.futures for Python applications
parallelism Using Docker for
containerization
Week 5
Code Quality, Testing,
Debugging
1 Code Quality (Writing Clean 2 Unit Testing (Writing Test Debugging Techniques
3
Code, Code Reviews) Cases,TDD) and Tools
Principles of clean code: readability,
Unit testing: importance, frameworks
maintainability, simplicity Debugging: strategies and
Code reviews: best practices, conducting (unittest, pytest)
Test-Driven Development (TDD): techniques
effective reviews
concepts, workflow Tools: using IDEs, debuggers,
logging
Machine Learning
Week 1
Intro and Supervised
learning basics
1 Introduction to ML 2 Supervised learning 3 Model validation and
What it is and why it is needed, KNN, for classification and evaluation
compare traditional “AI” against regression task
traditional softwar Approximate NN Data splitting, bias-variance
Paradigms of ML tradeoff, validation/evaluation,
metrics
Week 2
Simplest models and what the
“learning” process looks like

1 Supervised learning for 2 How to improve 3 Supervised learning for

regression hyperparameters, regularization, classification
grid search
Linear regressio Logistic regression vs Naive
algebraic interpretatio Baye
probabilistic interpretatio transforming text into numbers
polynomial fitting
Week 3
Practical issues in data
preparation
1 Data cleaning 2 Feature engineering 3 Feature scaling
Data cleaning, handling missing techniques and real examples min/max, max/abs,
values, encoding categorical transformations (z-score, log
values and etc)
Week 4
Practical issues cont’d and
other model families
1 Imbalanced datasets 2 Other families for 3 Supervised learning
what to do at different steps: data supervised learning wrap-up
preparation, training and metrics
SVM (SVC, SVR
Decision trees, random forests,
gradient boosting
Week 5
Unsupervised learning
1 PCA 2 Density estimation, 3 Clustering, Kmeans,
histograms, KDE hierarchical
Week 6

Unsupervised learning cont’d

and intro to R

1 DBScan GMM, EM Reinforcement learning

2 3

basics
Week 7
Review and Practical
Exercises

1 Review and Practical 2 Practical Session 3 Practical Session

Exercises on Machine (Applying Supervised (Applying
Learning Algorithms Learning Models) Unsupervised Learning
Review of key concepts and Models)
algorithms learned so far
 Deep Learning and
Common Applications
Week 1
Introduction to neural
networks

1 What are NNs, why DL 2 Basic MLP under 3 Learning with

got so popular the lens backpropogation

Issue of feature engineerin Derivation of update formula

MLE framework, forward
Cases for MSE and binary CE
General info about most common propagation, loss construction,
architectures (Perceptron-MLP, univariate regression, binary/
CNN, RNN, Transformers) multiclass classification,
multitask learning
Week 2

1 Activation functions 2 Initialization and 3 Regularization

What makes NN non-linea optimization Common and DL specific
Commonly used functions regularization methods
Gradient descent variations
Week 3
CNNs and intro to
transfer learning
1 CNN basics 2 CNN architectures, 3 Applications
CNNs on 1d inpu Transfer learning Object detection, segmentatio
CNNs on 2d inpu YOLO, Faster-RCNN, U-Net,
Downsampling and upsampling
basics Mask R-CNN
AlexNet, Inception, VGG, ResNe
Transfer learning with CNNs
Week 4
Transfer learning for
texts, RNNs
1 Text representation 2 How to encode 3 RNNs
Word, subword embeddings and sequential information Vanilla RN
how to learn the Vanishing/exploding gradient
Word2vec training, negative Logistic regression and MLP on LSTM, GRU, BiLSTM
sampling embedding
CNNs for texts
Week 5

RNNs cont’d

1 RNNs in practice Sequence generation Improved seq2seq

2 3

Sequence classification, sequence Introducing attention block for

MLP vs RNN approach for

labeling tasks seq2seq

language modelin

Seq2seq: Machine translation

Week 6
Attention is all you need!
1 Further improved 2 BERT, pretraining/ 3 Decoder-only setup,
seq2seq finetuning aka GPT
Birth of Transformer architectur Sequence classificatio Pretraining/finetunin
Building blocks Sequence pair classificatio GPT family, scalin
Sequence labeling Birth of new paradigm: in-
context learning
Week 7
Transformers for vision, multi-
modal learning, practical problems
1 Transformers in computer Multi-modal learning 3 Model compression
2
vision
Distillation and quantization
CLIP
Vision transformers, Masked autoencoders
Week 8
Guest Lectures
1 Learning on Graphs 2 RAG 3 Multi-modal learning
Review and
final project