AI & SaaS Glossary
90 key AI, machine learning, and SaaS terms explained in plain language.
A
A/B Testing
A/B testing is a method of comparing two or more versions (such as web pages, ads, or features) to determine which one performs better.
Agentic AI
Agentic AI refers to AI systems that can autonomously set goals, make plans, and carry out tasks — like an intelligent assistant that proactively thinks and acts.
AI Agent
An AI agent is an AI program that can perceive its environment, set goals on its own, and execute tasks to achieve them. Think of it as a digital assistant that not only answers questions but also proactively carries out a series of actions for you.
API
An API (Application Programming Interface) is like a standard menu, allowing one software program to securely and reliably request and use the features or data of another.
ARR
ARR (Annual Recurring Revenue) is a key metric for measuring the health of a SaaS (Software as a Service) company, representing the total stable revenue the company can expect to receive each year from all of its subscription customers.
Authentication
Authentication is the process of verifying the identity of a user or system, much like unlocking a door with a key or entering a password to log into an account.
Authorization
Authorization is the process by which a system or service verifies whether a user or program has permission to perform a specific action or access a specific resource.
B
B2B
B2B stands for Business-to-Business, referring to transactions of products or services between companies rather than directly to everyday consumers.
B2C
B2C (Business-to-Consumer) is a business model in which companies sell products or services directly to end consumers.
BBR
BBR (Bottleneck Bandwidth and Round-trip propagation time) is a TCP congestion control algorithm proposed by Google. Instead of treating packet loss as a congestion signal, it actively measures a link's bottleneck bandwidth and minimum round-trip time, pacing packets accordingly — significantly improving throughput and latency on high-latency, lossy paths.
Bootstrapped
Bootstrapped refers to a company or project that is launched and grown entirely with the founder's personal funds or early revenue, without relying on external investment.
C
CAC
CAC (Customer Acquisition Cost) is the average total cost a business spends to acquire one new customer.
CDN
A CDN (Content Delivery Network) is a network of edge nodes spread across the globe that serves static content — images, video, scripts — and even dynamic requests from locations near the user, reducing latency, offloading origin servers, and improving availability.
Chain of Thought
Chain of Thought is a technique that lets AI models show their step-by-step reasoning before giving a final answer, much like how humans list their thinking steps when solving a problem.
Churn Rate
Churn rate is the percentage of customers who stop using a product or service over a given period.
CI/CD
CI/CD is a software development approach that helps teams release software updates more frequently and reliably by automating the build, test, and deployment processes.
Computer Vision
Computer vision is the technology that enables computers to 'see' and understand image and video content the way humans do.
Context Window
A context window is the maximum length of conversation or text that an AI model (such as a chatbot) can consider and remember at once when generating a response.
Conversion Rate
Conversion rate is the percentage of visitors who complete a specific goal action, such as making a purchase, signing up, or clicking. It measures how effectively marketing or user experience turns visitors into customers or users.
Caching
Caching stores the result of a computation or request close to where it will be used so identical future requests can reuse it. On the web, caches exist at the browser, CDN edge, reverse proxy, application, and database layers — making it the highest-leverage performance optimization available.
D
DevOps
DevOps is a culture and working method that brings together software development (Dev) and IT operations (Ops) teams, aiming to build, test, and release software faster and more reliably through automation and collaboration.
Diffusion Model
A diffusion model is an AI technique that generates entirely new, realistic images, audio, or video by gradually adding noise to data and then learning how to reverse that process.
DNS
DNS (Domain Name System) is like the internet's phone book: it translates the website names you type in (such as www.example.com) into numeric addresses (IP addresses) that computers understand, allowing you to access websites.
Docker
Docker is an open-source platform that lets developers package an application and all of its dependencies into a lightweight, portable container, ensuring the app runs the same way in any computing environment.
F
Feature Flag
A feature flag is a software technique that lets developers dynamically turn specific application features on or off without redeploying code.
Few-shot
Few-shot learning is the ability of AI models to understand and perform new tasks from just a handful of examples.
Fine-tuning
Fine-tuning is the process of giving additional specialized training to an already-trained general-purpose AI model so it can better perform specific tasks or understand domain-specific knowledge.
Foundation Model
A foundation model is a large AI model pretrained on massive amounts of data. It gains broad knowledge and general capabilities that can serve as the basis for many specific tasks.
Freemium
Freemium is a business model that offers a free basic product to attract users while generating revenue from paid premium features or services.
Function Calling
Function Calling is the process by which a large language model (LLM) identifies the intent behind a user's request and outputs, in a predefined format, the parameters needed to invoke an external tool or function.
G
Generative AI
Generative AI is a type of artificial intelligence that creates entirely new content—such as text, images, music, or code—by learning patterns from existing data.
Git
Git is a version control system for tracking and managing file changes. Like a smart "time machine," it records every modification made during a project's development and makes it easy for teams to merge each person's work when collaborating.
GraphQL
GraphQL is a query language for APIs that lets clients specify exactly which data they need from the server, instead of fetching a fixed data structure as with traditional APIs.
I
IaaS
IaaS (Infrastructure as a Service) is a cloud computing model where you use resources like virtual servers, storage, and networking on demand — essentially renting virtual computing hardware instead of buying and maintaining physical equipment yourself.
Indie Hacker
An Indie Hacker is an entrepreneur who independently builds and runs their own software product or service, usually working alone or in a small team, with the goal of creating a profitable, sustainable business.
Inference
Inference is the process by which an AI model analyzes and evaluates new input based on what it has learned, and generates answers or predictions.
K
KPI
A KPI (Key Performance Indicator) is a specific, quantifiable measure of how well an organization, team, or individual is performing against its key objectives.
Kubernetes
Kubernetes is an open-source system for automating the deployment, management, and scaling of containerized applications. Think of it as an intelligent, automated robot butler dedicated to managing complex applications made up of many independent "containers".
L
Landing Page
A landing page is a standalone web page designed specifically to capture visitors and guide them toward a specific action, such as signing up, making a purchase, or downloading a file.
LLM
An LLM (large language model) is a powerful AI that can understand and generate human-like text, like a knowledgeable conversation partner.
Load Balancing
Load balancing is a technique that intelligently distributes network traffic or computing tasks across multiple servers, ensuring that no single server becomes overloaded, thereby improving the overall performance and reliability of the system.
Low-code
Low-code is a software development approach that allows developers to quickly build applications through graphical interfaces and minimal hand-written code.
LTV
LTV (Customer Lifetime Value) is the total revenue a customer is expected to generate for a company over the entire course of using its product or service.
M
Marketplace
A marketplace is an online platform that centrally showcases and sells a variety of software, services, or digital products, where users can browse, purchase, and integrate different tools or applications.
MCP
MCP (Model Context Protocol) is an open protocol initiated by Anthropic that standardizes how AI applications connect to external tools and data sources. Think of it as a USB port for AI: any MCP server can expose databases, ticketing systems, file systems, or internal APIs to MCP-capable AI clients in a uniform way.
Microservices
Microservices is a software architecture style that breaks a large, complex application into a series of small, independent services, each focused on a specific business function and communicating with one another through network interfaces (usually APIs).
Monorepo
A Monorepo (single repository) is a software development strategy in which the code for multiple related projects or services is kept in one version control repository.
MRR
MRR (Monthly Recurring Revenue) is the total predictable income a subscription-based SaaS company reliably receives each month from all of its active customers.
Multi-modal
Multi-modal refers to AI systems capable of processing and integrating multiple types of information inputs at the same time, such as text, images, audio, and video.
MVP
An MVP (Minimum Viable Product) is the leanest initial version of a product, containing only its most essential features, built to quickly validate market assumptions and gather user feedback.
N
nftables
nftables is the Linux kernel's packet filtering and NAT framework that replaces the long-serving iptables. It unifies syntax across IPv4, IPv6, ARP, and bridging, matches rules more efficiently through sets and maps, and supports transactional rule changes. It has shipped in mainline kernels since 3.13 and is the default firewall backend on current distributions.
NLP
NLP (Natural Language Processing) is a branch of artificial intelligence that enables computers to understand, interpret, and generate human language.
No-code
No-code is a software development approach that lets users build applications or automated workflows through graphical interfaces and drag-and-drop operations, without writing any traditional code.
NPS
NPS (Net Promoter Score) is a metric that measures customer loyalty and satisfaction by asking customers, "How likely are you to recommend our product or service to a friend or colleague?"
O
OAuth
OAuth is a secure authorization protocol that lets you grant an application (such as a social media management tool) access to your data on another service (such as your Twitter account) without sharing your password.
OKR
OKR (Objectives and Key Results) is a goal-management framework that helps teams and individuals set, track, and achieve ambitious goals.
Open Beta
An open beta is a testing phase in which a software product or service is opened to the general public before its official release, aiming to gather broad feedback and uncover potential issues.
Open Source
Open source is a development model in which a software's source code can be freely viewed, used, modified, and distributed by anyone.
P
PaaS
PaaS (Platform as a Service) is a cloud computing service that provides users with a ready-made platform for easily developing, running, and managing applications without having to worry about the underlying infrastructure.
PLG
PLG (Product-Led Growth) is a business model centered on making the product itself the main driver of acquiring, converting, and retaining customers.
PMF
PMF (product-market fit) means a product successfully meets a strong demand in a specific market, achieving sustained user growth and business returns.
Product Hunt
Product Hunt is an online platform for daily launches and discovery of new tech products, where users share, discuss, and vote for their favorite apps, websites, and hardware devices.
Product Roadmap
A product roadmap is a strategic document that clearly outlines a product's direction over a coming period and the features planned for release.
Prompt Engineering
Prompt engineering is the design and refinement of the questions or instructions given to AI models (especially large language models) to guide them toward more accurate, relevant, and useful answers.
R
RAG
RAG (Retrieval-Augmented Generation) is a technique that enables AI models to look up and cite up-to-date, accurate information from external knowledge bases when answering questions.
Reinforcement Learning
Reinforcement learning is a method that lets AI learn through trial and error, much like training a pet: rewards for doing things right, penalties for doing them wrong, until it learns to complete a specific task.
REST API
A REST API is a standard way for different software systems to communicate over the internet, using common web requests (such as retrieving, creating, updating, or deleting data) to exchange information.
Retention Rate
Retention rate is the percentage of users or customers who continue to use a product or service over a specific period of time, indicating how well a business keeps its audience engaged.
RLHF
RLHF (Reinforcement Learning from Human Feedback) is a method for training AI models that collects human preference feedback on model outputs to guide the model toward generating responses that better align with human values and expectations.
Reverse Proxy
A reverse proxy is an intermediary deployed on the server side: client requests hit the proxy first, which forwards them to backend services and returns the responses. Clients only ever see the proxy's address; the backend architecture stays completely hidden. Nginx, Caddy, HAProxy, and most cloud load balancers are reverse proxies.
Rate Limiting
Rate limiting protects services by capping the number of requests allowed in a time window: excess requests are rejected or queued. It prevents traffic spikes from overwhelming a system, and also powers anti-abuse and API metering. In high-availability design it is the core trade of sacrificing some requests to keep the whole service up.
S
SaaS
SaaS (Software as a Service) is a model of delivering software over the internet: instead of installing and maintaining it locally, users simply access it through a browser or app.
SDK
An SDK (Software Development Kit) is a set of pre-written code, tools, and documentation that helps developers build applications more easily for a specific platform or system.
Serverless
Serverless is a cloud computing model in which developers only write and upload code, while the cloud provider automatically handles server operation, scaling, and maintenance.
SLA
An SLA (Service Level Agreement) is a formal commitment between a service provider and a customer that clearly defines the performance standards the service must meet, such as uptime or response speed.
SSL/TLS
SSL/TLS is a security protocol that works like a lock on internet communications, ensuring the information you send and receive online (such as passwords and credit card numbers) can't be stolen or tampered with.
SSO
SSO (Single Sign-On) is an authentication method that lets users sign in once with a single set of credentials to access multiple related software applications or websites, without entering credentials for each service separately.
T
Token
In artificial intelligence, a token is the basic unit of text when processed by an AI model — it can be a word, a punctuation mark, or even part of a word.
Transformer
A Transformer is a deep learning model architecture for processing sequential data such as text. Using a technique called attention, the model can focus on all parts of the input at once, so it understands and generates content more effectively.
TypeScript
TypeScript is a superset of JavaScript that adds an optional static type system on top of JavaScript, letting developers catch errors early while writing code and improve maintainability.
V
Vector Database
A vector database is a database specialized in storing and quickly retrieving the mathematical vectors (strings of numbers) generated by AI models to represent the meaning of data.
Viral Loop
A viral loop is a product growth mechanism in which existing users naturally attract new users as they use the product, creating a self-reinforcing growth cycle.
VPS
A VPS (Virtual Private Server) is a product that splits one physical server into multiple isolated virtual machines through virtualization. Each VPS has its own operating system, an allocated share of CPU, memory, and disk, and its own public IP. With root access, you can run it like a dedicated server.
W
Webhook
A webhook is a mechanism that lets one application automatically send real-time notifications to another — like automatically 'phoning' the other app when an event occurs.
WebSocket
WebSocket is a web communication protocol that allows a persistent, bidirectional connection between a client (such as a browser) and a server, enabling real-time, efficient data exchange.
White-label
White-label refers to a company that produces a product or service but allows other companies to repackage and sell it under their own brand name.