AI knowledge map

Master the ideas behind modern AI.

Move from foundations to architectures, retrieval, agents, and production systems with crisp explanations tied directly to MachinoAI research.

ML

Machine Learning

Systems that learn patterns from data and use those patterns to make predictions, decisions, or recommendations.

Supervised learning maps inputs to known labels.Unsupervised learning discovers structure without explicit labels.

DL

Deep Learning

Neural-network based learning where layered representations turn raw data into increasingly useful abstractions.

Representation learning reduces the need for manual feature engineering.Backpropagation assigns credit through many layers.

LLMs

Large Language Models

Large neural networks trained on text and other modalities to predict, generate, reason over, and transform information.

Token prediction can produce broad language and reasoning abilities.Context windows define what the model can attend to in one request.

RAG

Retrieval-Augmented Generation

A pattern where a model retrieves relevant external knowledge before generating an answer.

Retrieval reduces dependence on memorized facts.Chunking and metadata shape what the model can cite.

Agents

AI Agents

AI systems that plan, call tools, observe results, and iterate toward a goal.

Tools extend a model beyond text generation.State and memory help the agent maintain context across steps.

MLOps

MLOps

The engineering practice of shipping, monitoring, governing, and improving machine learning systems in production.

Reproducibility makes experiments trustworthy.Monitoring catches data drift and performance regressions.