RAHULX.COM
Glossary · File GLS-001

Tech & AI Glossary

Plain-English definitions for the technology and artificial intelligence terms that show up everywhere but rarely get explained. No prior background assumed.

A

Algorithm
A finite, step-by-step set of instructions for solving a problem or completing a task, independent of the programming language used to implement it.
API (Application Programming Interface)
A defined contract that lets one piece of software request services or data from another without needing to know its internal implementation.
Artificial General Intelligence (AGI)
A hypothetical AI system that could match or exceed human ability across essentially any intellectual task, as opposed to today's systems, which excel at narrow, specific tasks.

B

Bandwidth
The maximum rate at which data can travel across a network connection, usually measured in bits per second.
Big Data
Datasets too large or complex for traditional data-processing tools to handle, typically characterized by volume, velocity, and variety.
Blockchain
A distributed, append-only ledger in which records ('blocks') are cryptographically linked in sequence, making past entries extremely difficult to alter without detection.

C

Cache
A small, fast storage layer that keeps copies of frequently used data close at hand so future requests can be served more quickly than fetching from the original source.
Cloud Computing
The delivery of computing resources — servers, storage, databases, networking — over the internet on demand, rather than owning and maintaining physical hardware.
Compute (AI context)
Shorthand for the processing power, usually measured in GPU/TPU hours or FLOPs, used to train or run a machine learning model.
Cryptography
The practice of securing information by transforming it so that only authorized parties can read or verify it, underpinning HTTPS, digital signatures, and cryptocurrencies alike.

D

Deep Learning
A subfield of machine learning that uses multi-layered neural networks to automatically learn patterns from large amounts of data, powering most modern AI systems including image recognition and large language models.
Diffusion Model
A generative AI technique that creates images or audio by starting from random noise and iteratively refining it into a coherent output, the approach behind tools like Midjourney and Stable Diffusion.

E

Edge Computing
Processing data physically closer to where it's generated (a device or local server) instead of sending everything to a distant cloud data center, reducing latency.
Encryption
The process of converting readable data into a coded form that can only be reversed with the correct key, protecting information in transit or at rest.

F

Fine-tuning
Additional training applied to an already-trained AI model using a smaller, task-specific dataset, adapting a general-purpose model to a narrower use case.
Firmware
Low-level software permanently programmed into a hardware device that controls its basic operation, sitting between the hardware and the operating system.

G

GPU (Graphics Processing Unit)
A processor originally designed for rendering graphics that turned out to be extremely efficient at the parallel matrix math used in training and running neural networks.

H

Hallucination (AI)
When an AI language model generates text that is fluent and confident-sounding but factually incorrect or entirely fabricated.

I

Internet of Things (IoT)
The network of everyday physical devices — thermostats, cameras, wearables, appliances — embedded with sensors and connectivity that let them collect and exchange data.

L

Large Language Model (LLM)
An AI model trained on vast amounts of text to predict and generate human-like language, the technology behind chatbots like ChatGPT and Claude.
Latency
The delay between a request being sent and a response being received, often the more noticeable performance factor for interactive applications compared to raw bandwidth.

M

Machine Learning (ML)
A branch of AI in which systems improve at a task by learning patterns from data, rather than following explicitly hand-coded rules.
Multimodal AI
AI systems that can process and generate more than one type of data — text, images, audio, video — within a single model, rather than being limited to one format.

N

Neural Network
A machine learning architecture loosely inspired by the brain, made of layers of interconnected 'neurons' that adjust connection strengths during training to recognize patterns.

O

Open Source
Software whose source code is made publicly available for anyone to view, use, modify, and redistribute, typically under a specific license governing those freedoms.

P

Phishing
A social-engineering attack that tricks people into revealing sensitive information or installing malware, usually via a message disguised as coming from a trusted source.

Q

Quantum Computing
A computing paradigm that uses quantum-mechanical phenomena such as superposition and entanglement to perform certain calculations far faster than classical computers, though still largely experimental for most real-world problems.

R

Robotic Process Automation (RPA)
Software 'bots' that mimic repetitive human actions within digital systems, such as copying data between applications, without requiring changes to the underlying software.

T

Token (AI/LLM context)
A chunk of text — often a word, part of a word, or punctuation mark — that a language model processes as its basic unit of input and output.
Two-Factor Authentication (2FA)
A login security method that requires two separate forms of verification, typically a password plus a one-time code sent to a device the user controls.

V

Virtual Private Network (VPN)
A service that routes internet traffic through an encrypted tunnel to a remote server, masking the user's IP address and shielding traffic from local network eavesdroppers.

Z

Zero-Day Vulnerability
A software security flaw that is discovered and potentially exploited before the developer has had any time ('zero days') to release a fix.