Technical Glossary

The iterative cycle of an LLM agent: receive input → decide on action (often a tool call) → execute → observe result → decide next action — until the agent reaches a final response.

An iterative approach to project management and software development that delivers work in small, incremental cycles with continuous feedback.

The field focused on ensuring AI systems behave as intended, don't cause harm, and remain aligned with human values and goals.

Single entry point that manages, secures, and routes API requests.

A unique identifier used to authenticate and authorize API requests, typically passed as a header or query parameter.

Standing up a fake version of an API that returns canned or configurable responses — lets frontend, mobile, or downstream service teams develop in parallel with the real backend or test edge cases that are hard to reproduce live.

Restricting the number of API requests a client can make in a given timeframe.

The practice of maintaining multiple versions of an API to allow gradual migration when breaking changes are introduced.

An automated process that routes a decision (purchase, refund, time off, contract) through one or more approvers based on rules — capturing the audit trail of who approved what and when.

Python's built-in async/await framework for writing concurrent, single-threaded I/O-bound code — the foundation for modern async Python web frameworks (FastAPI, Starlette, Sanic).

A neural network component that allows models to focus on the most relevant parts of the input when producing each part of the output.

A rule or algorithm that assigns conversion credit across the marketing touchpoints that influenced a customer's decision — 'how do we credit a signup that saw a Facebook ad, then clicked a Google ad, then opened an email?'

An immutable chronological log of every change made to a record or system — captures who changed what, when, and (ideally) why, supporting compliance, security, and forensic analysis.

ASG

An AWS construct (or equivalent on other clouds) that automatically scales the number of compute instances up or down based on demand metrics — ensuring capacity matches load without manual intervention.

A link from one website to another, serving as a 'vote of confidence' that signals content quality and authority to search engines.

An HTTP authorization scheme that passes a token in the request's Authorization header (`Authorization: Bearer <token>`) — the dominant pattern for API authentication, including OAuth 2.0 access tokens.

An opinionated, deterministic Python code formatter that reformats code to a consistent style with minimal configuration — eliminates style debates by giving developers no choices.

A deployment strategy that maintains two identical production environments, routing traffic to the new version only after it's verified healthy.

BM25

A classical lexical (keyword-based) ranking function for full-text search — counts term occurrences with diminishing returns and document-length normalization. Still the strongest non-neural baseline for search.

The percentage of visitors who leave a website after viewing only one page without taking any further action.

A program in which an organization pays security researchers for responsibly reporting vulnerabilities — turns the global researcher community into an extension of the internal security team.

BI

Technologies and strategies for analyzing business data to support decision-making.

BPA

Using technology to execute recurring tasks or processes in a business where manual effort can be replaced.

Software that executes business logic and decisions based on predefined rules, separating decision logic from application code.

An HTML element that tells search engines which version of a URL is the 'master' copy, preventing duplicate content issues.

A distributed task queue for Python that enables asynchronous processing and scheduled job execution.

Celery's task scheduler — runs periodic tasks on cron-like or interval schedules, dispatching them to Celery workers for execution.

CoT

A prompting technique that instructs an LLM to show its step-by-step reasoning before answering — dramatically improves accuracy on multi-step problems by giving the model space to think.

CDC

A pattern for streaming row-level changes (inserts, updates, deletes) from a source database to downstream systems in near-real-time — typically by tailing the database's write-ahead log.

COA

The complete listing of every account in an organization's general ledger, organized by category (assets, liabilities, equity, revenue, expenses).

CI/CD

Continuous Integration and Continuous Deployment - automated software delivery practices.

NetSuite's three native dimensions for slicing GL transactions beyond the Chart of Accounts — supporting segment-level reporting, budgeting, and analysis without account proliferation.

CTR

The percentage of people who click on a link after seeing it, commonly measured for search results, ads, and email campaigns.

An analytics technique that groups users by a shared attribute (usually signup date or first purchase) and tracks how each cohort's behavior evolves over time — separates trends from compositional shifts.

CV

AI systems that can interpret and understand visual information from images and video.

Decision rules in a workflow that branch behavior based on field values, prior steps, or external data — 'if X then do A, else do B' patterns that power dynamic, context-aware automation.

A lightweight, standalone package that includes everything needed to run a piece of software — code, runtime, libraries, and settings.

CDN

A distributed network of servers that delivers web content from locations geographically closest to each user.

A strategic marketing approach focused on creating and distributing valuable, relevant content to attract and retain a clearly defined audience.

Python's standard library module for context-local state — like thread-local variables but works correctly across asyncio tasks, where threading.local() doesn't.

CRO

The systematic process of increasing the percentage of website visitors who take a desired action (purchase, sign up, contact, etc.).

CWV

A set of Google metrics measuring real-world user experience: loading performance (LCP), interactivity (INP), and visual stability (CLS).

The number of pages a search engine will crawl on your site within a given timeframe, determined by crawl rate and crawl demand.

A time-based task scheduler in Unix/Linux systems that executes commands or scripts at specified intervals.

CORS

A browser security mechanism that controls which domains can make requests to your API, preventing unauthorized cross-origin access.

CSRF

An attack that tricks authenticated users into performing unintended actions on a web application they're logged into.

XSS

A web security vulnerability that allows attackers to inject malicious scripts into web pages viewed by other users.

A multidimensional analytics structure that pre-aggregates measures across many dimensions — designed for fast slice-and-dice queries that would be slow against raw fact tables.

A NetSuite-specific layout for transaction or entity records that lets you control field visibility, ordering, and which subtabs appear for different user roles or business units.

A SuiteScript-based extension that lets you customize how NetSuite posts to the General Ledger for specific transaction types — modifying or adding GL impact lines beyond what the standard accounting engine produces.

User-defined data structures in NetSuite for storing business-specific information.

A data structure that speeds up query performance by creating a fast lookup path to rows based on specific column values.

A version-controlled change to a database schema — adding tables, modifying columns, or creating indexes — that can be applied and rolled back consistently.

The process of organizing database tables to reduce data redundancy and improve data integrity by following normal form rules.

Horizontally partitioning data across multiple database instances, distributing load and enabling storage beyond single-server limits.

A Python 3.7+ decorator (@dataclass) that auto-generates __init__, __repr__, __eq__, and other boilerplate for plain-data classes — eliminates manual code for value-object patterns.

Automatically appending additional data to a record from external sources — adding firmographic data (employee count, industry, revenue) to a lead from just their email or company name.

Centralized repository storing structured and unstructured data at scale.

The process of creating a visual representation of data structures, relationships, and constraints for a database or system.

Automated workflow for moving and processing data from source to destination.

The process of converting data structures into a format suitable for transmission or storage, such as JSON, XML, or Protocol Buffers.

A centralized repository optimized for analytical queries, storing structured historical data from multiple sources.

dbt

A SQL-first data transformation framework that brings software engineering practices (version control, testing, documentation, modular models) to analytics workflows — the 'T' in ELT pipelines.

DDoS

Distributed Denial of Service — coordinated attack that floods a target service with traffic from many sources, overwhelming capacity until legitimate users can't connect.

The strategic adoption of digital technologies to fundamentally change how a business operates, delivers value, and competes.

A high-level Python web framework that provides an ORM, admin interface, authentication, and 'batteries-included' approach to rapid web development.

DRF

A powerful toolkit for building Web APIs in Django, providing serializers, viewsets, authentication, and browsable API documentation.

A metric (developed by Moz) predicting how likely a website is to rank in search results, scored from 1 to 100 based on backlink profile.

A series of pre-written emails (or other messages) sent automatically on a schedule — each message 'dripped' over time at a defined interval after a trigger event.

Processing data near the source rather than in centralized cloud servers.

Experience, Expertise, Authoritativeness, and Trustworthiness — Google's framework for evaluating content quality, especially for YMYL topics.

Network traffic leaving a network boundary — used in cloud billing to refer specifically to data leaving the cloud provider's network (often expensive) vs ingress (free in most clouds).

Numerical representations of text that capture semantic meaning for AI processing.

Protecting stored data by converting it into an unreadable format that can only be decrypted with the correct key.

A dynamic value stored outside the application code that configures behavior across different environments (development, staging, production).

ERP

Enterprise Resource Planning — integrated software that manages core business processes including finance, HR, manufacturing, supply chain, and CRM.

ETL

Extract, Transform, Load - process for moving data between systems.

EDA

System design where actions are triggered by specific events or state changes.

A modern, high-performance Python web framework for building APIs, with automatic OpenAPI documentation and async support.

A special search result displayed above organic results ('Position 0') that directly answers a user's query with extracted content from a web page.

Teaching AI models new tasks with just a few examples.

Adapting a pre-trained AI model to specific tasks or domains by training it on specialized data.

A production-ready animation library for React that provides declarative animations, gestures, and layout transitions.

An LLM capability that allows the model to invoke predefined functions or APIs as part of its response, enabling it to take real-world actions.

An analytics technique that measures user progression through a series of sequential steps (signup → activate → trial → convert → retain) — identifies where users drop off and what fixes would move the most volume.

GenAI

AI systems that can create new content including text, images, code, and data based on training patterns.

GA4

Google's web analytics platform for tracking website traffic, user behavior, conversions, and marketing performance.

GSC

Google's free tool for monitoring website performance in search, submitting sitemaps, and identifying indexing issues.

Query language and runtime for APIs that allows clients to request specific data.

A quantum search algorithm that finds a target item in an unsorted database in O(√N) operations vs the classical O(N) — a quadratic speedup for any problem expressible as 'search a space for an item satisfying some condition'.

gRPC

A high-performance, language-agnostic RPC framework built by Google — uses HTTP/2 for transport and Protocol Buffers for binary message serialization, far faster than REST/JSON for service-to-service communication.

A Python WSGI HTTP server for running Django and Flask applications in production, managing multiple worker processes.

When AI models generate false or nonsensical information that appears plausible.

An API endpoint that reports whether a service is running and able to serve requests, used by load balancers and monitoring systems.

Adding more machines or instances to handle increased load, distributing traffic across multiple servers.

A modern Python HTTP client that supports both sync and async APIs — drop-in replacement for requests, with HTTP/2 support and seamless asyncio integration.

Combining multiple automation technologies — AI, RPA, process mining, low-code — to automate as many business processes as possible.

A cloud IAM (Identity and Access Management) construct that bundles permissions and can be temporarily 'assumed' by a user or service — eliminates the need to embed long-lived credentials in code or configuration.

A property where performing the same operation multiple times produces the same result, preventing duplicate processing.

The process by which search engines add web pages to their database (index) so they can appear in search results.

The process of using a trained AI model to make predictions or generate outputs on new data — the 'production' phase of AI.

IaC

Managing infrastructure through machine-readable definition files.

IDP

Using AI to extract, classify, and process information from unstructured documents.

Links between pages on the same website that distribute page authority, establish content hierarchy, and help users and search engines navigate.

A NetSuite item record for physical goods you stock — tracking on-hand quantity, cost, valuation method (FIFO/LIFO/Average), bin/location, and reorder points across warehouses.

iPaaS

Integration Platform as a Service - cloud platforms for building and deploying integrations.

A visual workflow management method that uses boards and cards to track work items through stages from start to finish.

KPI

A measurable value that demonstrates how effectively an organization is achieving key business objectives.

The process of discovering and analyzing search terms that people enter into search engines, to inform content strategy and SEO targeting.

A model-training technique where a small 'student' model learns to mimic the outputs of a large 'teacher' model — producing a much smaller model that retains most of the teacher's quality.

K8s

Container orchestration platform for automating deployment and scaling.

LLM

An AI model trained on vast amounts of text data capable of understanding and generating human-like text.

An automated sequence of touchpoints (email, ads, content) that progressively educates and qualifies a prospect over weeks or months until they're ready to talk to sales.

A model that assigns numeric scores to leads based on fit (firmographic data) and engagement (behavioral signals) — used to prioritize sales outreach to the highest-probability buyers.

A system that distributes incoming network traffic across multiple servers to ensure no single server becomes overwhelmed.

Collecting, centralizing, and analyzing log data from multiple services and servers in a single searchable platform.

A specific, multi-word search phrase with lower search volume but higher conversion intent than broad keywords.

LoRA

A parameter-efficient fine-tuning method that adds small trainable rank-decomposed matrices to a frozen pre-trained model — drastically reduces fine-tuning compute and storage vs full fine-tuning.

Development platforms that enable building applications with minimal or no hand-written code through visual interfaces and drag-and-drop components.

A cloud offering where the provider operates, scales, patches, backs up, and monitors a software service (database, message queue, container runtime) so the customer only configures and uses it.

MDM

Ensuring consistency and accuracy of shared data across the organization.

An asynchronous communication mechanism that decouples systems by storing messages until the receiving system is ready to process them.

An HTML attribute providing a brief summary of a page's content, displayed as the snippet text in search engine results.

Architectural pattern breaking applications into small, independent services.

Software that connects two or more applications, translating data formats and managing communication between systems.

MVP

The simplest version of a product that delivers core value to early users and generates validated learning for further development.

MCP

An open protocol developed by Anthropic for connecting LLMs to external data sources, tools, and systems — standardizes how AI applications discover and call capabilities provided by separate MCP servers.

The maximum amount of text (measured in tokens) that a language model can process in a single request.

A technique for reducing AI model size and computational requirements by using lower-precision numbers (e.g., 8-bit instead of 32-bit).

The practice of collecting, analyzing, and acting on data about system health, performance, and behavior in real-time.

mTLS

A TLS handshake where both client and server present and verify certificates — proves the identity of BOTH parties, not just the server. Common in zero-trust networks and service-to-service authentication.

MFA

A security method requiring two or more verification factors (password + phone code, biometric, hardware key) to access an account.

AI systems that can process and generate multiple types of data — text, images, audio, video — in a unified model.

NetSuite's capability to manage multiple legal entities, brands, or business units within a single account with consolidated reporting.

A static type checker for Python that uses type hints (PEP 484) to verify type correctness without running code — catches whole classes of bugs at lint time.

NLP

The branch of AI focused on enabling computers to understand, interpret, and generate human language.

Full-stack React framework with server-side rendering and API routes.

NISQ

A term coined by John Preskill in 2018 describing the current era of quantum computing — devices with 50–10,000 qubits but without error correction, capable of demonstrating quantum effects but not yet running cryptographically-significant algorithms.

A category of databases that store data in non-tabular formats — documents, key-value pairs, graphs, or wide columns — optimized for specific access patterns.

Industry-standard protocol for authorization and API access delegation.

Optimization techniques applied directly to web page content and HTML source code to improve search engine rankings.

OpenAPI

A standard, language-agnostic interface description for HTTP APIs (formerly Swagger) — JSON or YAML document that describes endpoints, parameters, request/response schemas, and authentication. The contract for modern REST API design.

ORM

Object-Relational Mapping — a programming technique that lets you interact with a database using objects and methods instead of raw SQL.

A regularly updated list of the ten most critical web application security risks, published by the Open Web Application Security Project.

PEFT

An umbrella term for techniques that fine-tune large pre-trained models by updating only a small subset of parameters — LoRA, adapters, prefix tuning, and prompt tuning all qualify.

The structured NetSuite process that locks an accounting period after all transactions are recorded, reconciliations are complete, and financial statements are finalized.

A social-engineering attack that tricks users into revealing credentials, clicking malicious links, or executing payloads — typically via spoofed email but increasingly via SMS (smishing), voice (vishing), and other channels.

Python's default package installer that downloads and installs packages from the Python Package Index (PyPI).

A Python dependency manager and packaging tool that uses pyproject.toml for declarative configuration, deterministic resolution via poetry.lock, and integrated virtualenv management.

An advanced open-source relational database known for reliability, extensibility, and standards compliance, widely used in production applications.

A shareable bundle of saved HTTP requests organized into folders — Postman's primary unit of API documentation, testing, and team collaboration.

PoLP

A security principle that every user, service, or process should have only the minimum permissions required to perform its function — and nothing more. Limits blast radius when credentials are compromised.

Using data from IT systems to discover, monitor, and improve real business processes by analyzing event logs.

A prioritized list of features, enhancements, bug fixes, and technical debt that represents all planned work for a product.

Creating large numbers of search-optimized pages automatically using templates and structured data, targeting long-tail keywords at scale.

PWA

Web applications that provide native app-like experiences.

The practice of designing and optimizing input prompts to get desired outputs from AI models.

A security attack where malicious instructions are embedded in user input to manipulate an AI model into ignoring its system instructions.

A NetSuite transaction that records a commitment to buy from a vendor — items, quantities, expected receipt dates, and per-item costs that flow into AP, inventory receipts, and accruals.

A Python data validation library that uses type hints to validate, serialize, and document data structures.

An extension of Pydantic that loads application settings from environment variables, .env files, secrets backends, or CLI args — with full type validation and IDE autocomplete.

Python's most popular testing framework, supporting simple assertions, fixtures, parameterization, and plugin extensibility.

A computational model for quantum computing that consists of qubits, quantum gates applied to those qubits, and measurement operations — the quantum analog of a classical logic circuit.

A quantum phenomenon where two or more qubits become correlated such that the state of one instantly determines the state of the others, regardless of distance — Einstein famously called it 'spooky action at a distance'.

QEC

Techniques for protecting quantum information from decoherence and gate errors by encoding logical qubits across many physical qubits — analogous to but vastly more complex than classical error correction.

A unitary operation that manipulates qubit states — the quantum analog of classical logic gates (AND, OR, NOT). Combinations of gates implement quantum algorithms.

QKD

A cryptographic technique that uses quantum properties (typically photon polarization) to share a secret key between two parties with the guarantee that any eavesdropping attempt will be detected — provides information-theoretic security from the laws of physics.

A demonstration that a quantum computer can perform a specific computation faster than any classical computer — the milestone Google claimed in 2019 with Sycamore, though for a contrived benchmark task rather than a practical problem.

The basic unit of quantum information — a two-state quantum system (like an electron's spin or a photon's polarization) that can exist in a superposition of 0 and 1 rather than just one or the other.

The process of improving database query performance through better SQL, indexing, schema design, and execution plan analysis.

RAG

Retrieval-Augmented Generation - combining LLMs with external knowledge bases for more accurate responses.

Malicious software that encrypts a victim's files and demands payment (usually cryptocurrency) for the decryption key — one of the most damaging modern cybercrime categories.

RBAC

An authorization model where permissions are assigned to roles, and roles are assigned to users — instead of attaching individual permissions to each user, drastically simplifying access management at scale.

JavaScript library for building user interfaces with component-based architecture.

ReAct

An LLM agent pattern that interleaves reasoning steps with tool-use actions — 'Thought → Action → Observation → Thought → ...' — until the model arrives at a final answer.

Processing and analyzing data immediately as it's generated.

A long-lived OAuth 2.0 credential that the client uses to obtain new short-lived access tokens without re-prompting the user for credentials — separates session lifetime from token-leak blast radius.

RLHF

A training technique where AI models are refined based on human preferences and evaluations of their outputs.

RDBMS

A database that organizes data into tables with rows and columns, using relationships (foreign keys) to connect related data across tables.

A second-stage retrieval refinement that takes the top results from initial retrieval (often semantic search) and reorders them using a more accurate but slower model — usually a cross-encoder.

RI

A 1- or 3-year commitment to a specific amount of cloud compute capacity in exchange for a significant discount (typically 30-70%) vs on-demand pricing — trades flexibility for predictable savings.

A web design approach that makes pages render well across all device sizes and screen orientations.

REST

Representational State Transfer - an architectural style for building web services.

Custom REST APIs built in NetSuite using SuiteScript.

A team meeting held after each sprint or project to reflect on what went well, what didn't, and how to improve processes.

A data integration pattern that moves data FROM the data warehouse INTO operational tools (CRM, marketing platforms, sales tools) — the opposite of traditional ETL, which moves data from operational systems into the warehouse.

A server that sits in front of backend servers, forwarding client requests and returning responses on their behalf.

RPA

Software robots that mimic human actions to complete repetitive tasks across applications.

An automated routing pattern that distributes incoming leads, tickets, or tasks evenly across a pool of assignees — typically using rotation or load-balancing logic.

An extremely fast Python linter and formatter written in Rust — replaces flake8, isort, pylint, and Black with one tool, running 10-100× faster on the same codebase.

A pre-defined sequence of sales touches (emails, calls, LinkedIn messages, etc.) spread over days or weeks — designed to systematically work a prospect from cold contact to qualified opportunity.

A NetSuite transaction that records a customer's commitment to purchase — capturing items, quantities, pricing, fulfillment terms, and payment terms before fulfillment and invoicing.

SAML

Security Assertion Markup Language — an XML-based standard for exchanging authentication and authorization data between an identity provider (IdP) and a service provider (SP). The longstanding standard for enterprise SSO.

NetSuite's powerful reporting and data extraction tool.

A NetSuite feature that uses a Saved Search to identify a set of records, then applies a bulk field update (often with a formula) across all matching records in one operation.

Structured data vocabulary (schema.org) added to HTML that helps search engines understand page content and display rich results.

An Agile framework using fixed-length sprints, defined roles (Product Owner, Scrum Master, Team), and ceremonies to manage complex projects.

SEO

The practice of optimizing websites and content to rank higher in search engine results pages, driving organic (non-paid) traffic.

SERP

The page displayed by a search engine in response to a user query, containing organic results, ads, featured snippets, and other SERP features.

The practice of securely storing, distributing, and rotating sensitive credentials like API keys, database passwords, and encryption keys.

A search approach that retrieves results based on meaning rather than keyword overlap — typically uses vector embeddings to find documents semantically similar to a query.

Cloud computing model where providers manage infrastructure and automatically scale resources.

SSE

A simpler alternative to WebSocket for server-to-client streaming over standard HTTP — text-only, server-push only (no client-to-server stream), but works through proxies/firewalls and reconnects automatically.

SSR

Generating the HTML content of a page on the server before sending it to the browser, improving initial load time and SEO.

A quantum algorithm that factors large integers in polynomial time — exponentially faster than the best known classical algorithms. The basis for quantum's threat to RSA encryption.

SPA

A web application that loads a single HTML page and dynamically updates content without full page reloads.

SSO

An authentication scheme where a user logs in once to a central identity provider and gains access to multiple applications without re-entering credentials — reduces password sprawl while centralizing identity controls.

SCD

A dimensional modeling pattern for tracking how a dimension attribute (customer address, product price, employee manager) changes over time — preserving history for accurate point-in-time analytics.

A variant of the Star Schema where dimension tables are normalized into multiple related tables — trades query simplicity and JOIN performance for reduced storage and easier dimension hierarchy updates.

Two complementary approaches to information retrieval: sparse (lexical, keyword-based — BM25, TF-IDF) and dense (neural, embedding-based — vector search). Modern systems combine both for best results.

A cloud compute instance offered at 50-90% discount vs on-demand pricing — but the provider can reclaim it with little notice (typically 2 minutes) when capacity is needed elsewhere.

A fixed time period (usually 1-4 weeks) during which a Scrum team works to complete a set of committed backlog items.

Python's de facto SQL toolkit and ORM — supports both a low-level expression language (Core) and a high-level object-relational mapper (ORM), with deep support for PostgreSQL, MySQL, SQLite, and most major databases.

An attack that inserts malicious SQL code into application queries, potentially exposing or modifying database contents.

A digital certificate that enables encrypted HTTPS connections between browsers and web servers, verifying server identity.

Any person or group with an interest in or influence over a project's outcome, including sponsors, users, developers, and leadership.

SOP

A documented, repeatable set of step-by-step instructions for executing a recurring business process — the foundation for consistent execution, training, and automation.

A dimensional data modeling pattern for analytics warehouses — central fact tables (numeric measurements) surrounded by denormalized dimension tables (descriptive context) like a star.

Patterns and tools for managing shared data (state) across components in a frontend application.

SSG

Pre-rendering all pages at build time into static HTML files, resulting in the fastest possible page loads.

A structured logging library for Python that emits machine-parseable log events (JSON, key-value) instead of free-text — the foundation for log aggregation, search, and alerting.

A subdivision of a VPC's IP address range — typically used to segment public-facing resources from private ones, and to distribute resources across multiple availability zones for resilience.

A discrete business entity within NetSuite OneWorld that maintains its own currency, accounting books, taxes, and legal-entity hierarchy under a single shared NetSuite account.

NetSuite's business intelligence and reporting platform.

A package of NetSuite customizations — scripts, workflows, custom records, custom fields, saved searches, forms — that can be installed, versioned, and updated across multiple NetSuite accounts.

NetSuite's comprehensive cloud development platform.

NetSuite's native e-commerce platform that provides integrated online storefronts connected directly to ERP data.

NetSuite's visual workflow automation tool for creating business process workflows without coding.

NetSuite's JavaScript-based customization platform for creating custom business logic.

NetSuite's SOAP-based web services API for integrating external applications.

A fundamental quantum phenomenon where a system exists simultaneously in multiple states until measured — a qubit in superposition is both 0 and 1 with associated probabilities, collapsing to one definite value upon observation.

A machine learning approach where models learn from labeled training data — input-output pairs that teach the model the correct mapping.

An open-source, browser-based interface that renders an OpenAPI spec as interactive, try-it-yourself API documentation — the de facto standard for API docs since 2011.

A utility-first CSS framework that provides low-level utility classes for building custom designs directly in HTML markup.

The accumulated cost of shortcuts, workarounds, and deferred maintenance in a codebase that makes future changes more difficult and risky.

The practice of optimizing a website's infrastructure and architecture to help search engines crawl, index, and render content effectively.

The basic unit of text that LLMs process, typically representing parts of words or punctuation.

TBA

NetSuite's secure authentication method for external integrations — uses OAuth 1.0a tokens instead of usernames/passwords, eliminating password rotation and reducing the attack surface for API access.

The structured definition of a function (name, description, JSON-schema parameters) that an LLM can call — the contract that lets the model decide when and how to invoke external capabilities.

A content strategy that organizes pages into groups around a central pillar page, with related subtopic pages linking back to it.

Applying knowledge gained from training on one task to a different but related task, dramatically reducing training time and data requirements.

A neural network architecture that uses self-attention to process sequential data, forming the backbone of all modern large language models.

Python syntax for declaring expected types of variables, function parameters, and return values, enabling static analysis and better tooling.

Typed superset of JavaScript that compiles to plain JavaScript.

A machine learning approach where models discover patterns and structure in data without labeled examples.

A short, simple description of a feature from the perspective of the end user, following the format: 'As a [user], I want [goal] so that [benefit].'

An extremely fast Python package installer and resolver written in Rust, designed as a drop-in replacement for pip, pip-tools, and virtualenv.

A lightning-fast ASGI server for running async Python frameworks like FastAPI and Starlette in production.

VQE

A hybrid quantum-classical algorithm that finds the minimum eigenvalue of a Hamiltonian — the workhorse for near-term quantum chemistry and material science applications, designed to be NISQ-friendly.

A specialized database optimized for storing and querying high-dimensional vectors (embeddings) for similarity search.

An isolated Python environment that contains its own packages and dependencies, preventing conflicts between projects.

VPC

A logically isolated section of a cloud provider's network where you launch resources with full control over IP ranges, subnets, route tables, and network gateways — your own private network in the public cloud.

A modern frontend build tool that provides instant dev server startup and optimized production builds using native ES modules and Rollup.

A11y

Designing and developing websites that are usable by people with disabilities, following WCAG guidelines.

WAF

A security system that monitors, filters, and blocks malicious HTTP traffic to and from a web application.

HTTP callbacks that notify external systems when events occur.

Two approaches to detecting changes: webhooks push notifications instantly when events occur, while polling checks for changes at regular intervals.

A bidirectional, full-duplex communication protocol over a single long-lived TCP connection — enables real-time server-to-client push (chat, live dashboards, collaborative editing) without HTTP polling.

NetSuite's native automation engine that triggers actions on record events — approvals, field updates, emails, and custom logic — without writing SuiteScript.

Coordinating and managing automated tasks and processes across multiple systems and teams.

A file that lists all important URLs on a website, helping search engines discover and prioritize content for crawling and indexing.

A no-code automation platform that connects 6,000+ apps through trigger-action workflows called 'Zaps.'

AI model's ability to perform tasks without specific training examples.

A security model that requires strict identity verification for every person and device trying to access resources, regardless of network location.