Glossary
The AI product development
The terms founders and product teams run into when building software with AI: PRD, MVP, RAG, MCP, TAM/SAM/SOM and more. Defined in plain English, in one place.
Jump to
A
6 terms- Acceptance criteria
- Acceptance criteria are the specific, testable conditions a feature must satisfy before it counts as complete, giving engineers, testers, and AI agents a shared, verifiable definition of done for each user story or requirement.
- Agent configuration
- Agent configuration is the set of parameters and instructions that define what an AI agent should do, how it should behave, which tools and skills it has access to, and when it should act independently versus deferring to a human.
- Agentic AI
- Agentic AI is a class of AI system that pursues a goal across multiple steps on its own: it plans an approach, calls tools to act, checks the results, and adapts, instead of returning one answer to one prompt.
- Agentic workflow
- An agentic workflow is a process in which AI agents carry a task from goal to finished output, choosing the steps and tools themselves at each stage, while humans set the objective, supply constraints, and approve the results.
- AI agent
- An AI agent is a software system that uses a language model to plan and execute multi-step tasks toward a goal, calling tools, checking results, and adjusting its approach without step-by-step human instructions.
- API
- An API (application programming interface) is a defined contract that lets one piece of software request data or actions from another, without either side needing to know how the other works internally.
B
2 terms- Backlog
- A backlog is the ordered list of features, bug fixes, improvements, and ideas a product team intends to build, ranked so that the items at the top represent the team's actual next commitments rather than a wish list.
- Bring your own keys (BYOK)
- Bring your own keys (BYOK) is a usage model that routes an AI tool's requests through your own API keys from an LLM provider, so usage bills to your account and your data flows under your own provider agreement.
C
2 terms- Code ownership
- Code ownership is the right to fully export, host, modify, and keep the source code a platform generates for you, so your product continues to exist independently of any vendor or subscription.
- Context window
- A context window is the maximum amount of text, measured in tokens, that an AI model can process in a single request, covering the system prompt, conversation history, documents, and the model's own response.
D
3 terms- Deployment
- Deployment is the process of building an application and publishing it to hosting infrastructure so real users can access it, typically behind your own domain with SSL and automated redeploys on every code change.
- Deployment / endpoint
- A deployment or endpoint is a hosted, callable instance of an AI model accessible via an API, where the model runs on the inference provider's infrastructure and receives requests from your application.
- Design system
- A design system is a reusable library of components, design tokens, and usage rules that keeps a product's interface visually and behaviorally consistent as it grows across screens, features, and teams.
E
1 term- Embedding
- An embedding is a numeric vector representation of text, images, or other content that captures semantic meaning, letting software measure similarity between items and power semantic search, retrieval, and recommendations.
F
1 term- Fine-tuning
- Fine-tuning is a training process that adapts a pre-trained AI model to a specific task, tone, or domain by continuing training on your own labeled examples, changing the model's weights rather than its inputs.
G
3 terms- Gateway / router
- A gateway or router is an abstraction layer that sits between an application and its AI model providers, routing each request to an appropriate model based on task type, cost, latency, and availability.
- Generation / completion
- Generation or completion is the output an AI model produces in response to a prompt, built token by token from the most statistically plausible next token given everything in the context window.
- GitHub sync
- GitHub sync is a live connection between a code-generating tool and a GitHub repository you own, so every generated change lands in your version control automatically, with full history and reviewable diffs.
H
1 term- Hallucination
- A hallucination is an AI output that states false or invented information with the same fluency and confidence as fact, such as citations that do not exist, functions that were never real, or statistics with no source.
I
3 terms- Inference
- Inference is the process of running a request through a trained AI model to produce an output, covering the full path from tokenizing the input prompt through generating and returning the completion.
- Inference provider
- An inference provider is a company that hosts AI models on its own GPU infrastructure and exposes them through an API, handling the scaling, availability, and billing so your application can call completions without managing servers.
- Integration
- An integration is a working connection between your product and an external service, such as payments, authentication, or email, so that data and actions flow between the two systems automatically.
J
1 term- JSON
- JSON (JavaScript Object Notation) is a text-based format for structuring and exchanging data, defined by RFC 8259, that represents values and objects using human-readable name-value pairs and ordered lists rather than code syntax.
K
1 term- Knowledge base
- A knowledge base is a curated, searchable repository of organized information that a product team or organization maintains to answer questions, preserve decisions, and provide consistent context to both human team members and AI systems.
L
1 term- Large language model (LLM)
- A large language model (LLM) is an AI model trained on massive text datasets to predict and generate language, powering writing, coding, analysis, and reasoning tools through token-by-token text generation.
M
7 terms- Market research
- Market research is the process of validating a product idea against real-world evidence, including market size, competitor gaps, pricing, and user needs, before building, making it the cheapest point in the product lifecycle to discover you are wrong.
- Minimum viable product (MVP)
- A minimum viable product (MVP) is the smallest version of a product that delivers real value to real users and produces validated learning about whether the core idea deserves further investment.
- Model
- A model is a trained AI system that predicts and generates text, code, or other content based on patterns learned during training, exposed through an API as a component that applications call to produce completions.
- Model Context Protocol (MCP)
- The Model Context Protocol (MCP) is an open standard that connects AI models to external tools and data sources through one consistent interface, so any compatible agent can discover and use a service without custom integration code.
- Model developer
- A model developer is an organization that designs, trains, and owns the weights of an AI model, defining its capabilities, safety behavior, and the core intelligence that inference providers then serve.
- Model family
- A model family is a group of related AI models released under one name by a single developer, sharing a core architecture and training approach but differing in size, capability, speed, and price.
- Multi-agent orchestration
- Multi-agent orchestration is the coordination of several specialized AI agents working toward one larger goal, using a routing layer that assigns each subtask to the right agent and carries shared context across every handoff.
N
2 terms- Native mobile app
- A native mobile app is an application built to run directly on iOS or Android, distributed through the App Store or Google Play, with full access to device capabilities such as push notifications, camera, and offline storage.
- North star metric
- A north star metric is the single measurement that best captures the core value a product delivers to its users, chosen so that sustained growth in that one number reflects genuine business progress rather than vanity activity.
O
1 term- Observability
- Observability is the ability to understand a software system's internal state solely from its external outputs, including structured logs, distributed traces, and metrics, so that distributed failures and anomalies can be diagnosed without guessing or reproducing them locally.
P
4 terms- Product requirements document (PRD)
- A product requirements document (PRD) is a structured specification that defines what a product or feature should do and why, covering the problem, target user, scope, user stories, and acceptance criteria that guide design and engineering.
- Prompt
- A prompt is the instruction given to an AI model that specifies the task, context, constraints, and output format the model should follow, forming the entire interface between your intent and the model's response.
- Prototype
- A prototype is an interactive model of a product, ranging from clickable design frames to a coded demo with sample data, built to test flows and assumptions with users before committing to production code.
- Provider
- A provider is a company that makes AI models accessible via an API, operating the infrastructure and managing the commercial relationship so applications can call completions without owning model weights or GPU servers.
Q
1 term- Quality assurance (QA)
- Quality assurance (QA) is a planned, systematic set of activities that gives confidence a software product and its development process meet specified requirements, covering the entire lifecycle from requirements through release rather than focusing solely on testing the finished build.
R
2 terms- Retrieval-augmented generation (RAG)
- Retrieval-augmented generation (RAG) is an AI technique that retrieves relevant documents from a knowledge base and inserts them into a language model's context at generation time, so answers are grounded in real, current data instead of training memory.
- RICE scoring
- RICE scoring is a prioritization framework that ranks product ideas by multiplying Reach, Impact, and Confidence, then dividing by Effort, producing a single comparable score that shows which backlog items deliver the most value per unit of work.
S
2 terms- Skill
- A skill is a packaged AI capability that extends an agent with a specific, defined function, such as searching the web, running code, or querying a database, usually exposed through one or more tools.
- System prompt
- A system prompt is a higher-priority instruction set given to an AI model before any user message that defines the model's role, rules, tone, and boundaries for an entire session or product.
T
3 terms- TAM, SAM & SOM
- TAM, SAM, and SOM are three nested market-size estimates, total addressable market, serviceable available market, and serviceable obtainable market, used to size an opportunity from the entire category down to the share a company can realistically win.
- Token
- A token is the basic unit of text that AI models read and generate, roughly four characters or three-quarters of an English word; model pricing, context windows, and generation speed are all measured in tokens.
- Tool
- A tool is a callable function an AI model invokes at runtime to act on the world, such as searching the web, reading files, or calling an API, extending what the model can do beyond generating text.
U
2 terms- User persona
- A user persona is a semi-fictional profile of a target user, assembled from research, that captures their role, goals, context, and frustrations so product decisions stay grounded in a real audience.
- User story
- A user story is a short, plain-language description of a feature told from the user's perspective, typically written as "As a [user], I want [goal] so that [benefit]," keeping development focused on outcomes rather than outputs.
V
1 term- Vector database
- A vector database is a storage system that indexes and retrieves high-dimensional vector embeddings, enabling similarity search by finding the nearest matches to a query vector in semantic space rather than by exact keyword matching.
W
1 term- Wireframe
- A wireframe is a low-fidelity layout of a screen that shows structure, hierarchy, and element placement using simple boxes and placeholder text, so teams can agree on flow before investing in visual design.
X
1 term- XAI (explainable AI)
- XAI (explainable AI) refers to methods and techniques that make the decisions and outputs of AI systems understandable to humans, so that the reasons behind a prediction or generation are inspectable rather than opaque.
Y
1 term- YAML
- YAML (YAML Ain't Markup Language) is a human-friendly data serialization language designed for configuration files and data exchange, using indentation and plain text to represent structured data without the syntax overhead of code-like formats.
Z
1 term- Zero-shot prompting
- Zero-shot prompting is a technique that gives an AI model a task and instructions without providing worked examples, relying entirely on the model's pre-trained knowledge to understand and execute the task from the instruction alone.
Stop defining the work
Turn any idea into a validated brief, a real PRD, a design, and production web and mobile code, on one AI platform.