Learn AI &amp; ML the Easy way

Modern ML

Attention

Focus a token and watch how strongly it draws on every other one, in the graph and in the matrix at once. Then watch the weights get spent: the token’s output is the blend they buy, and the temperature decides whether it says anything.

Foundations

Bias-Variance Tradeoff

Sweep a polynomial from a flat underfit to a wild overfit and watch validation error bottom out and turn back up. Then change how much data the model gets, and watch the same degree go from reckless to reasonable.

Evaluation & RL

Threshold, Confusion Matrix & ROC

Drag one threshold and watch the confusion matrix, precision, recall and the ROC point all move together. Then push it to either extreme and see one metric go perfect while the other collapses.

Modern ML

Transformer Decoding

A language model hands you a distribution, not a word. Step through generation and watch greedy decoding walk itself into a loop, then use temperature and truncation to get variety without nonsense.

Supervised Learning

Gradient Descent

Drag the data, drag the line, then watch an optimiser find the fit for you — one real gradient step at a time. Push the learning rate too far and watch it fly apart.

Visuals

Core ML

Classical ML, deep learning, optimisation, and neural network foundations — taught through numbered SVG sequences

Computer Vision

Classical image processing, geometry, detection, and vision architectures — taught through numbered SVG sequences

Concepts

Gen AI

Transformers, LLMs, RAG, Agents, and the generative AI stack

Core ML

Classical ML, deep learning, training patterns, and foundations

Data Concepts

Data engineering, pipelines, and data-centric AI

Math

Linear algebra, calculus, probability for ML

NLP Concepts

Classic NLP pipeline, POS tagging, NER, sentiment analysis, and text processing

Papers

Attention Is All You Need

The landmark 2017 paper that introduced the Transformer architecture, replacing RNNs with self-attention and launching the modern era of large language models.

LLaMA 1

LLaMA: Open and Efficient Foundation Language Models

The 2023 paper from Meta that introduced the first highly capable, open-weights foundation model, sparking the open-source LLM revolution.

Llama 3

The Llama 3 Herd of Models

The 2024 Meta technical report detailing a massive scale-up in data and compute, proving that dense models can reach frontier-level capabilities through sheer scale and data quality.

Llama 2

Llama 2: Open Foundation and Fine-Tuned Chat Models

The 2023 paper from Meta that introduced a commercially viable, chat-tuned foundation model, detailing the immense effort required for RLHF and safety alignment.

GPT-4

GPT-4 Technical Report

The 2023 technical report from OpenAI detailing the model that defined the frontier of AI capabilities, demonstrating human-level performance on professional benchmarks.

Interview Questions

"How does an agent decide when to ask the user for clarification versus act?"

Quick answer

Agents decide to ask for clarification through explicit instructions, threshold-based uncertainty logic, or by providing them a specific 'AskUser' tool that they can invoke when lacking critical parameters.

"How would you design agent-to-agent communication and shared state without conflicts?"

Quick answer

Design communication via a centralized state graph (like LangGraph) where agents pass structured messages, or use a blackboard pattern where agents read/write to a shared memory object using strict schemas.

"How would you design guardrails and an audit trail for an agent operating in a regulated domain?"

Quick answer

Design guardrails using dual-LLM validation, deterministic output filtering, mandatory human-in-the-loop for state changes, and an immutable logging system capturing every prompt, thought, and tool execution.

"How would you design permissions and least-privilege access for agents that take real-world actions?"

Quick answer

Implement least privilege by scoping agent credentials using OAuth/IAM, creating specialized agents with limited toolsets, utilizing human-in-the-loop approvals, and logging all actions immutably.

"What does the agent loop (observe, think, act) look like?"

Quick answer

The agent loop is a continuous cycle where the agent observes its environment, reasons about the current state, decides on an action, and executes it, repeating until the goal is met.

Cheat Sheets

Core Machine Learning

Problem framing, generalization, classic supervised and unsupervised models, features, validation, and leakage.

Data Structures and Algorithm Patterns

Complexity, core patterns with Python templates, graphs, dynamic programming, and the cues that point to each pattern.

Generative AI

LLM fundamentals, transformers, decoding, RAG, agents, fine-tuning, inference cost, evaluation, and production safety.

Math for Machine Learning

Linear algebra, calculus, optimization, probability, information theory, and the numerics that keep models stable.

Model Evaluation Metrics

Confusion-matrix metrics, ROC and PR curves, calibration, thresholds, multiclass averaging, regression, ranking, and reporting.

Learning Paths

Machine Learning Engineer

For developers who want to build production ML. Go from Python and math foundations to training, deploying, and monitoring classical, deep learning, and LLM systems end to end.

Data Scientist

For aspiring data scientists who want to turn raw data into decisions. You will be able to analyze data, build and evaluate ML models, communicate insights, and ship models to production.

Computer Vision Engineer

For developers who want to build vision systems. Learn image processing, deep learning and 3D vision, then train, optimize and deploy models on cloud and edge hardware.

AI Research Engineer

For engineers aiming at frontier AI research. Build math, ML and systems depth, then train, align, scale and evaluate models and publish reproducible results.

Data Analyst

For beginners and career switchers who want to turn raw data into decisions. Learn spreadsheets, SQL, BI, Python and statistics, then build a portfolio and land an analyst role.