A reference from Weave

Engineering & AI glossary

Understand the metrics, models, and methods behind modern engineering. Clear definitions, practical examples, and a closer look at what the numbers actually mean.

All terms

2,008 terms

  • AI ROI attribution

    AI ROI attribution is connecting measured benefits and costs to a particular AI capability. It gives teams a way to name, measure, or reason about an economic property of an AI workload without treating raw usage as proof of value.

    Token costs and AI ROI

  • AI ROI baseline

    AI ROI baseline is the documented starting point used to compare an AI investment. It gives teams a way to name, measure, or reason about an economic property of an AI workload without treating raw usage as proof of value.

    Token costs and AI ROI

  • AI ROI confidence

    AI ROI confidence is a statement of how certain an AI return estimate is and why. It gives teams a way to name, measure, or reason about an economic property of an AI workload without treating raw usage as proof of value.

    Token costs and AI ROI

  • AI scenario analysis

    AI scenario analysis is comparison of distinct future operating cases for an AI workload. It gives teams a way to name, measure, or reason about an economic property of an AI workload without treating raw usage as proof of value.

    Token costs and AI ROI

  • AI sensitivity analysis

    AI sensitivity analysis is testing how an ROI estimate changes when assumptions move. It gives teams a way to name, measure, or reason about an economic property of an AI workload without treating raw usage as proof of value.

    Token costs and AI ROI

  • AI showback

    AI showback is a transparent report of AI consumption and cost without an internal transfer. It gives teams a way to name, measure, or reason about an economic property of an AI workload without treating raw usage as proof of value.

    Token costs and AI ROI

  • AI spend forecast

    AI spend forecast is an estimate of future AI expense using usage, price, and workload assumptions. It gives teams a way to name, measure, or reason about an economic property of an AI workload without treating raw usage as proof of value.

    Token costs and AI ROI

  • AI spend forecast error

    AI spend forecast error is the difference between projected AI spend and incurred expense. It gives teams a way to name, measure, or reason about an economic property of an AI workload without treating raw usage as proof of value.

    Token costs and AI ROI

  • AI Test Generation

    AI Test Generation is a software-engineering concept describing how an AI coding system, its tools, or human collaborators handle a defined task.

    AI coding and agents

  • AI time saved

    AI time saved is working time avoided or redirected by an AI-assisted workflow. It gives teams a way to name, measure, or reason about an economic property of an AI workload without treating raw usage as proof of value.

    Token costs and AI ROI

  • AI total cost of ownership

    AI total cost of ownership is the full cost of operating AI across providers, infrastructure, people, integration, and governance. It gives teams a way to name, measure, or reason about an economic property of an AI workload without treating raw usage as proof of value.

    Token costs and AI ROI

  • AI usage attribution

    AI usage attribution is the practice of assigning requests and tokens to people, products, or workflows. It gives teams a way to name, measure, or reason about an economic property of an AI workload without treating raw usage as proof of value.

    Token costs and AI ROI

  • AI volume discount

    AI volume discount is a price reduction tied to reaching a stated usage level. It gives teams a way to name, measure, or reason about an economic property of an AI workload without treating raw usage as proof of value.

    Token costs and AI ROI

  • AI-assisted development

    AI-assisted development is software work that uses an AI system to help with activities such as explaining code, generating or editing code, writing tests, reviewing changes, or navigating a repository.

    AI coding and agents

  • AI-assisted software factory

    An AI-assisted software factory is a software delivery system in which coding assistants, generative tools, or agents participate in development activities alongside human teams. Its performance depends on the surrounding feedback, review, platform, governance, and production systems, not only on how much code AI produces.

    Measurement and experimentation

  • Alert deduplication

    Alert deduplication is a reliability concept used to describe a specific condition, control, or decision in the operation of software services.

    Reliability and observability

  • Alert evaluation window

    Alert evaluation window is a reliability concept used to describe a specific condition, control, or decision in the operation of software services.

    Reliability and observability

  • Alert fatigue

    Alert fatigue is the reduced ability or willingness to respond carefully to alerts after repeated exposure to notifications that are noisy, low priority, or rarely actionable.

    Reliability and observability

  • Alert inhibition

    Alert inhibition is a reliability concept used to describe a specific condition, control, or decision in the operation of software services.

    Reliability and observability

  • Alert policy

    Alert policy is documented conditions, grouping, suppression, ownership, and response expectations for alerts.

    Reliability and observability

  • Alert routing

    Alert routing is sending an alert to the team, channel, or escalation path responsible for response.

    Reliability and observability

  • Alert rule

    Alert rule is the query, condition, window, grouping, and action that create an alert.

    Reliability and observability

  • Alert suppression

    Alert suppression is the intentional prevention or grouping of notifications when known conditions make individual alerts redundant.

    Reliability and observability

  • Alert threshold

    Alert threshold sets the boundary for notification or response.

    Engineering analytics

  • Alert to action time

    Alert to action time is a software delivery concept used to describe a specific event, interval, control, or operating condition in the path from source change to production behavior. A useful definition names the boundary, unit, and decision the measure supports.

    DORA and DevOps

  • Alpha testing

    Alpha testing is a software testing or test-design practice used to gather evidence about a defined risk, behavior, boundary, or operating condition. It makes the question under test explicit, identifies the inputs and observations that matter, and gives a team a repeatable basis for deciding whether the result is acceptable.

    Code quality and technical debt

  • Alternative hypothesis

    Alternative hypothesis is a statistical or measurement concept used to describe, compare, or interpret engineering data. Its meaning depends on the unit of analysis, data-generating process, and question being asked.

    Measurement and experimentation

  • Anomaly detection

    Anomaly detection is a statistical or measurement concept used to describe, compare, or interpret engineering data. Its meaning depends on the unit of analysis, data-generating process, and question being asked.

    Measurement and experimentation

  • ANOVA

    ANOVA is a statistical or measurement concept used to describe, compare, or interpret engineering data. Its meaning depends on the unit of analysis, data-generating process, and question being asked.

    Measurement and experimentation

  • Anti-corruption layer

    Anti-corruption layer is a software maintenance concern describing a condition that can make future changes, verification, operation, or ownership harder. Its practical importance depends on supported behavior, rate of change, and the consequences of delay.

    Code quality and technical debt

  • API churn

    The frequency or magnitude of changes to a public interface over a defined period.

    Code quality and technical debt

  • API gateway for LLMs

    API gateway for LLMs is a model-routing or gateway concept used to manage interface consistency for AI requests. It describes a distinct decision, control, interface, or observation point between an application and one or more model providers.

    Model routing and gateways

  • API surface area

    The amount of public functionality that a component exposes to external callers.

    Code quality and technical debt

  • API testing

    API testing is a software testing or test-design practice used to gather evidence about a defined risk, behavior, boundary, or operating condition. It makes the question under test explicit, identifies the inputs and observations that matter, and gives a team a repeatable basis for deciding whether the result is acceptable.

    Code quality and technical debt

  • API versioning

    API versioning is a software maintenance concern describing a condition that can make future changes, verification, operation, or ownership harder. Its practical importance depends on supported behavior, rate of change, and the consequences of delay.

    Code quality and technical debt

  • Application monitoring

    Application monitoring is observation of service-level errors, latency, dependencies, and business operations.

    Reliability and observability

  • Apprenticeship

    Apprenticeship is a way to organize, support, or evaluate software work so that teams can make useful progress with less avoidable friction. It is most valuable when connected to a concrete outcome and the local conditions of the team using it.

    Developer productivity

  • Approval

    Approval is a practical concept in a pull-request workflow that shapes how people examine, discuss, own, or integrate a proposed change.

    Code review

  • Approval bottleneck

    An approval bottleneck occurs when a required reviewer or approval rule constrains the flow of otherwise ready changes.

    Code review

  • Approval churn

    Approval churn is the frequency with which approvals become stale, are dismissed, or need to be repeated after a change is updated.

    Code review

  • Approval condition

    Approval condition is a practical concept in a pull-request workflow that shapes how people examine, discuss, own, or integrate a proposed change.

    Code review

  • Approval dismissal

    Approval dismissal is the removal of an approval from the set of decisions that currently permits a change to merge.

    Code review

  • Approval expiration

    Approval expiration means a prior approval no longer satisfies the current merge policy, often because the change or target branch changed.

    Code review

  • Approval gate

    An approval gate is a workflow checkpoint that prevents a change from advancing until a defined reviewer decision is recorded.

    Code review

  • Approval latency

    Approval latency is the time between a change becoming reviewable and the required approval being recorded.

    Code review

  • Approval policy

    An approval policy states who must review which changes and what evidence is needed before integration.

    Code review

  • Approval rate

    Approval rate is the proportion of review requests or pull requests that receive the required approval within a defined population and period.

    Code review

  • Approximate nearest neighbor

    Approximate nearest neighbor is a language-model concept about generation behavior and sampling. It names a mechanism, representation, prompting pattern, decoding control, or context behavior that can change how an AI system processes input and produces output.

    LLM fundamentals

  • Architectural boundary

    A defined separation between components that limits responsibilities, dependencies, or change impact.

    Code quality and technical debt

  • Architecture churn

    The rate at which major structural relationships or boundaries change over time.

    Code quality and technical debt