Context Engineering
The practice of shaping what information an AI model sees—ensuring it has the right data, instructions, and constraints to produce relevant, accurate outputs.
Dataset
A structured collection of data used to train or evaluate an AI system. The quality and relevance of the dataset directly impact how well the AI performs.
Deep Learning
A type of machine learning that uses layered neural networks to identify complex patterns in large volumes of data, often powering advanced capabilities like language and image understanding.
Fine-Tuning
The process of further training a pre-trained model on specific data to make it more accurate and relevant for a particular use case or domain.
Foundation Model
A large, general-purpose AI model trained on broad data that can be adapted to a wide range of tasks, from text generation to analysis.
Guardrails
The rules, controls, and constraints put in place to ensure AI operates safely, compliantly, and within defined boundaries.
Hallucination
When an AI generates information that sounds credible but is incorrect, unverified, or fabricated, especially when not grounded in trusted data.
Inference
The process of using a trained model to generate outputs or predictions. Every interaction with an AI system (e.g., asking a question) is an inference.
Large Language Model (LLM)
A type of AI model trained on vast amounts of text data to understand and generate human language. Examples include GPT, Claude, and Gemini.
Machine Learning (ML)
A subset of AI where systems learn patterns from data over time, improving performance without being explicitly programmed for every scenario.
Model
The trained system itself is a mathematical representation that takes inputs (like text or data) and produces outputs (like predictions or generated content).
Neural Network
A computational model inspired by the human brain, made up of interconnected layers that process and transform data to recognize patterns or make predictions.
Parameters (Weights)
The internal numerical values within a model that are adjusted during training. These determine how the model interprets inputs and generates outputs.
Prompt
The input provided to an AI system—such as a question, instruction, or context—that guides the response it generates.
Prompt Engineering
The practice of structuring prompts to improve the quality, accuracy, and usefulness of AI outputs.
Reinforcement Learning (RL)
A training approach where models learn through trial and error, using feedback (rewards or penalties) to improve performance over time.
Supervised Learning
A machine learning approach where models are trained on labeled data, learning to map inputs to known outputs.
Token
The basic unit of text an AI model processes—typically a word or part of a word. Models read and generate language one token at a time.
Training
The process of teaching a model by feeding it data and adjusting its parameters so it can perform a specific task more effectively.
Unsupervised Learning
A machine learning approach where models identify patterns or groupings in data without predefined labels.
Vibe Coding
A prompt-driven approach to software development where a user describes what they want, and AI generates the code—often iteratively and conversationally.
