Artificial Intelligence23 Mar 202518 min read

The AI & ML Resource Library

A curated, no-fluff library of the articles, courses, and tools I recommend for going from AI beginner to confident practitioner.

1. Start Here

Artificial Intelligence can feel overwhelming at first. You know how it is.

You've probably heard that AI is dominated by complex maths and advanced statistics.. making it accessible only to seasoned academics. And yes, that's partly true, but AI today is different.

The good news? AI isn't reserved just for PhDs anymore. In fact, it's never been more approachable or practical to learn, even if math isn't your strong suit.

Don't get me wrong; groundbreaking research still requires top-tier researchers to advance the field.

If that weren't the case, we wouldn't have:

Titans: Learning to Memorize at Test Time – Google's alternative to the standard transformer architecture.
Mamba: Linear-Time Sequence Modeling - Making extremely long sequences efficient and powerful.
Frontier Models are Capable of In-context Scheming – when models follow misaligned goals.
Large Physics Models – Researching the development of Large Physics Models.

But that's advanced stuff. You're here because you want practical, hands-on experience—quick wins that take you from beginner to confident practitioner as smoothly as possible.

That's precisely why this resource exists. You'll find only the most essential online articles, blogs, courses, and so on — no unnecessary theory, no fluff. Detailed theory can wait for later. Here, the focus is purely practical.

The materials you find here will help you develop skills in Artificial Intelligence, data science, and machine learning.

Basics First

People learn differently. Some of the more common cases:

Creating a new GitHub repo and breaking stuff till it works.
Picking up a book and starting from theory
Mixed approach with articles, books, courses, and hands-on stuff.

This material isn't intended just for developers.

Articles

A beginner's guide to artificial Intelligence and machine learning
What is artificial Intelligence (AI)? – Best for new learners. Includes videos, too.
Companies are suddenly declaring themselves "AI first." Why it's a problem for their current customers. – This was way back in 2018. Wow.
https://pair.withgoogle.com/guidebook/patterns - This is a must-read if you are a fan of Human-Centered AI.
Building Domain-specific AI Apps https://cloud.google.com/blog/products/ai-machine-learning/three-step-design-pattern-for-specializing-llms
https://design.google/library/people-ai-research

Books

Перспективата изкуствен интелект – In Bulgarian, must read for anyone who wants to know AI – From Experts Systems, Deep Learning, and Computer Vision to LLMs. It's the only good Bulgarian book on the topic (to my knowledge).
Applying Data Science: How to Create Value with Artificial Intelligence – AI & Data Science without the hype and focus on the coding aspect.
Artificial Intelligence Basics: A Non-Technical Introduction
Co-Intelligence: Living and Working with AI – The title speaks for itself.

Videos

AI, Machine Learning, Deep Learning and Generative AI Explained - Non-Technical, Semi-Technical, Technical
All Machine Learning algorithms explained in 17 min - Non-Technical, Semi-Technical, Technical.
Generative AI in a Nutshell - how to survive and thrive in the age of AI - Non-Technical, Semi-Technical, Technical
[1hr Talk] Intro to Large Language Models - Semi-Technical, Technical
Let's build GPT: from scratch, in code, spelled out. - Technical
Let's build the GPT Tokenizer. - Technical
How I use LLMs - Non-Technical, Semi-Technical, Technical

Courses

AI For Everyone – Timeless classic, recommended as a first course to everyone.
Generative AI for Everyone – same as above but for GAI.
ChatGPT Prompt Engineering for Developers – Useful for everyone, not just developers.
The Data Science Course: Complete Data Science Bootcamp 2025 - This technical and hands-on course is helpful for semi-technical and non-technical people. It includes hands-on coding while explaining Data Science algorithms' business and technical value.
Machine Learning, Data Science and Generative AI with Python – Best hands-on ML, DS, and Generative AI using Python and TensorFlow. Recommended for technical folks.
Ultimate AWS Certified AI Practitioner AIF-C01 – For the AWS folks.
AI-900 Microsoft Azure AI Fundamentals – For the Azure folks.
The Product Management for AI & Data Science Course is excellent for Business Analysts and Designers.
Elements of AI—This is a great course that will help you genuinely understand AI and all its merits in a fun and interactive way.
Building AI – The more advanced piece can be taken by non-technical and technical people alike.

Want the Hands on? GitHub Projects

Artificial Intelligence for Beginners is hands-down the best resource for technical people. It offers a great mix of theory, study, and real-life projects.
AgentStack – Creating your very first AI Agent.

1.1 AI Beginner's Reading List

Intro

The list below is almost a direct copy of Jack Soslow’s AI Reading List, for which he takes full credit. I have only added some links that I believe are incredibly beneficial in their respective tiers.

S Tier (“Best of the Best”)

Tim Urban (2015) – The AI Revolution: The Road to Superintelligence

Andrej Karpathy (2017) – Software 2.0

Gwern (2020) – The Scaling Hypothesis

Wikipedia – The History of Artificial Intelligence

Google Research, Vaswani et al. (2017) – Attention Is All You Need

Transforming Ideas into Reality: How AI Fuels My Productivity & Creativity – As practical as it gets. Great for techies and non-techies alike.

A Tier

効 SynthLang, a hyper-efficient prompt language inspired by Japanese Kanji cutting token costs by 90%, speeding up AI responses by 900%

Eliezer Yudkowsky (2001) – Creating Friendly AI Summary

Tim Urban (2015) – The AI Revolution: Our Immortality or Extinction

DeepMind (2017) – AlphaGo

Eliezer Yudkowsky (2017) – There’s no Fire Alarm for Artificial Intelligence

Jay Alammar (2018) – The Illustrated Transformer

OpenAI (2020) – Language Models are Few-Shot Learners (GPT-3)

DeepMind (2020) – MuZero: Mastering Go, Chess, Shogi, and Atari without rules

OpenAI (2020) – AI and Efficiency

OpenAI (2021) – CLIP: Connecting Text and Images

Nostalgebraist (2022) – Chinchilla’s Wild Implications

Gwern (2022) – It looks like you’re trying to take over the world

Best of the Rest

Scott Alexander (2016) – Superintelligence FAQ

OpenAI (2017) – Learning to Communicate

OpenAI (2017) – Evolution Strategies as an Alternative to Reinforcement Learning

OpenAI (2017) – Learning to Cooperate, Compete, and Communicate

OpenAI (2017) – Proximal Policy Optimization

OpenAI (2017) – Competitive Self-Play

OpenAI (2018) – AI and Compute

DeepMind (2019) – Capture the Flag: The Emergence of Complex Cooperative Agents

OpenAI (2019) – OpenAI Five Defeats Dota 2 World Champions

OpenAI (2019) – Deep Double Descent

Jay Alammar (2020) – How GPT-3 Works: Visualizations and Animations

Evhub (2020) – 11 Proposals for Building Safe Advanced AI

DeepMind (2022) – Discovering Novel Algorithms with AlphaTensor

DeepMind (2020) – Using JAX to Accelerate Our Research

DeepMind (2020) – AlphaFold: A Solution to a 50-Year-Old Grand Challenge in Biology

Leigh Marie Braswell (2021) – Startup Opportunities in Machine Learning Infrastructure

Meta AI (2021) – Teaching AI How to Forget at Scale

OpenAI (2021) – DALL-E: Creating Images from Text

OpenAI (2021) – Multimodal Neurons in Artificial Neural Networks

OpenAI (2021) – Improving Language Model Behavior by Training on a Curated Dataset

OpenAI (2021) – OpenAI Codex

OpenAI (2022) – Aligning Language Models to Follow Instructions

OpenAI (2022) – DALL-E 2

OpenAI (2022) – Learning to Play Minecraft with Video PreTraining

OpenAI (2022) – Introducing Whisper

OpenAI (2022) – ChatGPT: Optimizing Language Models for Dialog

Nature (2022) – What’s Next for AlphaFold

Meta AI (2022) – CICERO: An AI Agent That Negotiates, Persuades, and Cooperates with People

Yann LeCun (2022) – How to Make AI Systems Learn and Reason Like Animals and Humans

Robert May (2022) – The Mental Model Most AI Investors Are Missing

Roon (2022) – Text Is the Universal Interface

Introducing Meta Agents: An agent that creates agents.

1.2 LinkedIn People to Follow

The AI field is far too dynamic.

All the information you knew last week is old news—not irrelevant, but not current. Take, for example, DeepSeek; since then, we have had new GPT models, a new Mistral model, LeChat, open-source data processing by DeepSeek, and Manus… And it hasn’t even been that long.

So, if you are serious about AI, I urge you to follow some people on LinkedIn. Their listing order doesn’t indicate their level of expertise, only the list update time.

Contact me if you want to be excluded from the list, and I will remove you.

2. Proposed AI Curriculum

Intro

This “curriculum” should not be taken as a regular curriculum would. The idea is to present 80% of the field with some follow-up notes and mentions so anyone interested can delve deeper into the field. I suggest you keep the list in mind when considering what area you might want to focus on next if you are serious about AI.

MODULE 1: Introductory Foundations

1.1 Mathematics Foundations

Subtopics

Linear Algebra Vectors, matrices, and tensors Matrix multiplication, determinants, eigenvalues/eigenvectors Singular Value Decomposition (SVD)
Calculus Multivariable calculus: partial derivatives, gradients, Hessians Chain rule in backpropagation
Probability & Statistics Probability distributions (discrete & continuous) Bayes’ theorem, conditional probability Expected values, variance, covariance Common distributions: Gaussian, Bernoulli, Binomial, Poisson
Discrete Mathematics Set theory, combinatorics Graph theory (basics, adjacency matrices)

Prerequisites

High-school level mathematics
Basic familiarity with algebra and trigonometry

Recommended Resources

Books: Linear Algebra and Its Applications by Gilbert Strang Calculus by James Stewart Probability and Statistics by Morris H. DeGroot & Mark J. Schervish
Online Courses: Khan Academy (Linear Algebra, Calculus, Probability) MIT OpenCourseWare (18.06 for Linear Algebra)
Tools: Python libraries: NumPy, SciPy

1.2 Computer Science Foundations

Subtopics

Programming Essentials Python fundamentals: data structures (lists, dictionaries), control flow, functions, OOP Familiarity with scripting, version control (Git)
Data Structures & Algorithms Arrays, stacks, queues, linked lists, trees, graphs Sorting and searching algorithms (Merge sort, Quick sort, Binary search) Complexity analysis (Big-O notation)
Introduction to Software Engineering Modular code design, testing, debugging Best practices in code development and deployment

Prerequisites

Basic programming experience (in any language)

Recommended Resources

Books: Introduction to Algorithms by Cormen, Leiserson, Rivest, and Stein Automate the Boring Stuff with Python by Al Sweigart
Online Courses: Harvard CS50 (Intro to Computer Science) Udemy / Coursera Python courses
Tools: Python 3, Git, GitHub or GitLab

MODULE 2: Classical AI

2.1 Historical Context & Foundations

Subtopics

History of AI Turing test, symbolic AI, Good Old-Fashioned AI (GOFAI)
AI Problem Solving Problem formulation, state-space representation, search trees

Prerequisites

Basic understanding of data structures and algorithms

Recommended Resources

Books: Artificial Intelligence: A Modern Approach (AIMA) by Stuart Russell & Peter Norvig (Chapters 1-3)
Key Papers: Computing Machinery and Intelligence by Alan Turing (1950)

2.2 Search & Heuristics

Subtopics

Uninformed Search Depth-first search (DFS), Breadth-first search (BFS), Uniform-cost search
Informed / Heuristic Search A*, Greedy best-first search, heuristic functions
Adversarial Search Minimax, alpha-beta pruning
Constraint Satisfaction Problems (CSPs) Backtracking, local search

Prerequisites

Familiarity with graph theory and complexity analysis

Recommended Resources

Books: Artificial Intelligence: A Modern Approach (Chapters 3-5)
Online Exercises: Implement BFS, DFS, A*, Minimax for simple games (like Tic-Tac-Toe)

2.3 Knowledge Representation & Expert Systems

Subtopics

Knowledge Representation Paradigms Logical representations (propositional logic, first-order logic) Semantic networks, frames, ontologies
Expert Systems Rule-based systems, inference engines Case studies: MYCIN, DENDRAL

Prerequisites

Familiarity with formal logic (propositional and predicate logic)

Recommended Resources

Books: Knowledge Representation and Reasoning by Ronald Brachman & Hector Levesque Expert Systems: Principles and Programming by Joseph Giarratano & Gary Riley
Key Papers: On the Criteria To Be Used in Decomposing Systems into Modules by D. L. Parnas (for design principles)

MODULE 3: Machine Learning Fundamentals

3.1 Overview of ML Paradigms

Subtopics

Supervised Learning Classification vs. regression Bias-variance tradeoff
Unsupervised Learning Clustering (k-means, hierarchical) Dimensionality reduction (PCA, t-SNE)
Reinforcement Learning Agent-environment interaction, reward maximization Markov Decision Processes (MDPs)

Prerequisites

Mastery of the foundational mathematics from Module 1
Familiarity with basic Python libraries (NumPy, pandas)

Recommended Resources

Books: Machine Learning by Tom M. Mitchell Pattern Recognition and Machine Learning by Christopher M. Bishop
Online Courses: Coursera: Machine Learning by Andrew Ng Stanford CS229 (open courseware)

3.2 Supervised Learning in Depth

Subtopics

Regression Linear regression, polynomial regression Regularization (Ridge, Lasso)
Classification Logistic regression, SVMs, Decision trees, Random forests Performance metrics (accuracy, precision, recall, F1-score, ROC AUC)
Ensemble Methods Bagging, boosting (AdaBoost, XGBoost)
Model Evaluation & Hyperparameter Tuning Cross-validation, grid search, Bayesian optimization

Recommended Resources

Books: An Introduction to Statistical Learning by James, Witten, Hastie, Tibshirani
Tools: scikit-learn, XGBoost

3.3 Unsupervised Learning in Depth

Subtopics

Clustering K-means, hierarchical clustering, DBSCAN
Dimensionality Reduction PCA, SVD, LDA, t-SNE, UMAP
Anomaly Detection Isolation Forest, Gaussian Mixture Models

Recommended Resources

Books: Hands-On Unsupervised Learning with Python by Ankur Patel
Tools: scikit-learn (clustering and decomposition modules)

3.4 Reinforcement Learning (Basics)

Subtopics

Markov Decision Processes (MDPs) States, actions, rewards, transition probabilities
Dynamic Programming Policy iteration, value iteration
Basic RL Algorithms Q-learning, SARSA

Recommended Resources

Books: Reinforcement Learning: An Introduction by Sutton & Barto
Online Courses: DeepMind’s RL Course on UCL’s website

MODULE 4: Deep Learning & Neural Networks

4.1 Neural Network Fundamentals

Subtopics

Perceptrons & MLPs Activation functions (ReLU, sigmoid, tanh, etc.)
Backpropagation & Optimization Gradient descent, stochastic gradient descent (SGD), momentum, Adam
Regularization Dropout, batch normalization, weight decay

Prerequisites

Understanding of calculus (gradients), linear algebra
Familiarity with machine learning fundamentals

Recommended Resources

Books: Deep Learning by Goodfellow, Bengio, and Courville
Online Courses: deeplearning.ai (Andrew Ng’s Deep Learning Specialization on Coursera)

4.2 Convolutional Neural Networks (CNNs)

Subtopics

Conv Layers & Pooling Filter kernels, stride, padding
Famous Architectures LeNet, AlexNet, VGG, ResNet, Inception
Applications Image classification, object detection (R-CNN, YOLO, SSD), segmentation (UNet)

Recommended Resources

Key Papers: ImageNet Classification with Deep Convolutional Neural Networks by Krizhevsky et al. (AlexNet) Deep Residual Learning for Image Recognition by He et al. (ResNet)
Tools: PyTorch, TensorFlow, Keras

4.3 Recurrent Neural Networks (RNNs)

Subtopics

Vanilla RNNs Sequence modeling, exploding/vanishing gradients
LSTM & GRU Gates (input, forget, output), cell state Applications in NLP, time series
Sequence-to-Sequence Models Encoder-decoder architectures (e.g., machine translation)

Recommended Resources

Key Papers: Long Short-Term Memory by Hochreiter & Schmidhuber (1997) Learning to Forget: Continual Prediction with LSTM by Gers et al.
Tools: PyTorch’s nn.LSTM, TensorFlow’s tf.keras.layers.LSTM

4.4 Training & Optimization Best Practices

Subtopics

Optimization Techniques Learning rate scheduling, adaptive methods Batch normalization
Model Tuning & Debugging Identifying overfitting vs. underfitting Monitoring loss curves, early stopping
Scalability & Distributed Training Data parallelism, model parallelism GPU, TPU training

Recommended Resources

Online Course: Fast.ai (Practical Deep Learning for Coders)
Tools: Horovod, Distributed TensorFlow

MODULE 5: Generative AI & Large Models

5.1 Generative Adversarial Networks (GANs)

Subtopics

GAN Basics Generator-discriminator framework Adversarial loss
Key Architectures DCGAN, WGAN, CycleGAN
Applications Image synthesis, style transfer

Prerequisites

Solid understanding of neural network architectures and training

Recommended Resources

Key Papers: Generative Adversarial Nets by Goodfellow et al. (2014) Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks by Radford et al. (DCGAN)
Tools: PyTorch, TensorFlow (GAN implementations)

5.2 Transformers & Attention Mechanisms

Subtopics

Sequence Transduction Self-attention, multi-head attention Positional encoding
Transformers Architecture Encoder-decoder stack, feed-forward layers, layer normalization
Key Breakthroughs BERT, GPT series, T5

Recommended Resources

Key Papers: Attention Is All You Need by Vaswani et al. (2017) BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding by Devlin et al. Language Models are Unsupervised Multitask Learners (GPT-2) by OpenAI
Tools: Hugging Face Transformers library

5.3 Diffusion Models & Other Generative Approaches

Subtopics

Diffusion Models Denoising diffusion probabilistic models (DDPM) Latent diffusion (e.g., Stable Diffusion)
VAEs (Variational Autoencoders) KL divergence, reparameterization trick
Applications Text-to-image (DALL·E, Stable Diffusion), text-to-text (ChatGPT)

Prerequisites

Familiarity with probabilistic modeling, advanced neural network concepts

Recommended Resources

Key Papers: Denoising Diffusion Probabilistic Models by Ho et al. Auto-Encoding Variational Bayes by Kingma & Welling
Tools: Hugging Face Diffusers library

5.4 Large Language Models (LLMs)

Subtopics

Pre-training & Fine-tuning Language modeling objectives (causal vs. masked) Fine-tuning for specific tasks
Prompt Engineering Zero-shot, one-shot, few-shot prompting Instruction tuning
Applications & Challenges Conversational AI (ChatGPT), summarization, code generation Hallucinations, alignment, interpretability

Prerequisites

Strong foundation in transformers and NLP

Recommended Resources

Key Papers: GPT-3: Language Models are Few-Shot Learners by Brown et al. Scaling Laws for Neural Language Models by Kaplan et al.
Tools: OpenAI API, Cohere, Anthropic LangChain for prompt engineering & advanced LLM workflows

MODULE 6: Ethics, Fairness & Bias in AI

6.1 Societal & Ethical Considerations

Subtopics

Data Bias & Algorithmic Bias Sources of bias, fairness metrics
Privacy & Data Protection GDPR, differential privacy
Explainability & Interpretability SHAP, LIME, feature importance

Recommended Resources

Books: Weapons of Math Destruction by Cathy O’Neil Fairness and Machine Learning by Barocas, Hardt, & Narayanan (online book)
Key Papers: A Framework for Understanding Unintended Consequences of Machine Learning by Suresh & Guttag

6.2 Governance & Regulation

Subtopics

Policy & Regulation AI Act (EU), Algorithmic Accountability Corporate governance frameworks
AI Safety & Alignment Alignment problem, interpretability vs. black-box models Long-term risk considerations (AGI, existential risks)

Recommended Resources

Organizations: AI Now Institute, Partnership on AI OpenAI on policy, DeepMind on safety
Papers: Concrete Problems in AI Safety by Amodei et al.

MODULE 7: AI & ML Applications

7.1 Natural Language Processing (NLP)

Subtopics

Traditional NLP Techniques Tokenization, stemming, lemmatization N-grams, TF-IDF
Deep NLP Word embeddings (Word2Vec, GloVe) Transformer-based NLP
Applications Chatbots, sentiment analysis, machine translation, question answering

Recommended Resources

Books: Speech and Language Processing by Jurafsky & Martin
Tools: NLTK, spaCy, Hugging Face Transformers

7.2 Computer Vision

Subtopics

Image Processing & Feature Extraction Edge detection, SIFT, SURF
Deep Vision CNN-based detection & segmentation
Applications Autonomous vehicles, medical imaging, facial recognition

Recommended Resources

Books: Computer Vision: Algorithms and Applications by Richard Szeliski
Frameworks: OpenCV, PyTorch, TensorFlow

7.3 Robotics & Control

Subtopics

Robotic Perception & Sensor Fusion LiDAR, camera-based SLAM (Simultaneous Localization and Mapping)
Reinforcement Learning in Robotics Policy-based and value-based methods Sim-to-real transfer
Motion Planning & Control Kinematics, trajectory optimization

Recommended Resources

Key Papers: DART: Optimizing a Whole-Body Controller for a Humanoid Robot via Differentiable Simulation
Tools: ROS (Robot Operating System), Gazebo

7.4 Finance, Healthcare & Other Domains

Subtopics

Finance Algorithmic trading, risk modeling, fraud detection
Healthcare Medical image diagnosis, drug discovery Personalized treatment recommendations
Edge AI & IoT Resource-constrained inference Embedded systems

Recommended Resources

Case Studies: JP Morgan’s AI-based trading algorithms Google DeepMind’s healthcare applications

MODULE 8: Advanced AI Research & Theoretical Frontiers

8.1 Advanced Topics in ML

Subtopics

Meta-Learning Model-agnostic meta-learning (MAML) Few-shot and zero-shot learning
Federated Learning Privacy-preserving collaborative training Challenges: data heterogeneity, communication overhead
Graph Neural Networks (GNNs) Message passing, graph embeddings (GraphSAGE, GCN)

Recommended Resources

Key Papers: Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks by Finn et al. Semi-Supervised Classification with Graph Convolutional Networks by Kipf & Welling
Tools: PyTorch Geometric, Deep Graph Library (DGL)

8.2 Theoretical Foundations & Interpretability

Subtopics

Learning Theory PAC learning, VC dimension Generalization bounds
Interpretability Research Saliency maps, feature visualization Neural network dissection
Causality in AI Structural causal models (SCMs) Interventions vs. correlations

Recommended Resources

Books: The Elements of Statistical Learning by Hastie, Tibshirani, Friedman (theoretical ML aspects) Causal Inference in Statistics: A Primer by Judea Pearl
Labs: Distill.pub (for interpretability research)

MODULE 9: Hands-on Projects & Real-World Implementation

9.1 Project Proposals & Case Studies

Project Ideas

Image Classification Web App

Build a CNN model and deploy as a web service (Flask/FastAPI).

CNNs, deployment, Docker, AWS/Heroku

NLP Chatbot

Fine-tune a Transformer for Q&A or conversation.

Transformers, Hugging Face, prompt engineering

GAN-based Image Generation

Create a GAN to generate stylized images (DCGAN/WGAN).

GANs, PyTorch/TensorFlow

Time Series Forecasting

Use RNNs or LSTMs for financial or demand forecasting.

RNNs, ARIMA baseline comparisons

Reinforcement Learning in a Simulated Environment

Implement Q-learning or PPO for a simple OpenAI Gym environment.

RL algorithms, Gym, stable-baselines3

Federated Learning Demo

Simulate federated learning on distributed data for privacy use-case.

Federated averaging, PySyft or TensorFlow Federated

Project Guidelines

Design: Clearly outline objectives, collect relevant data, define success metrics.
Implementation: Use version control, write modular code, systematically track experiments.
Evaluation: Use appropriate metrics, compare with baselines, and document performance.
Deployment: Containerize using Docker; consider cloud deployment (AWS, GCP, Azure).

aimachine-learningresourceslearning