ES-100 Glossary¶
Purpose¶
This glossary defines the shared language used in ES-100 and across the ETIS Engineering Platform.
Consistent terminology reduces cognitive load. Engineers should not have to guess whether different words mean the same thing.
Core platform terms¶
ETIS¶
Engineering Trustworthy Intelligent Systems.
ETIS is a framework for building, governing, operating, and stewarding intelligent systems in the AI era.
ETIS Engineering Platform¶
The repository-centered implementation of ETIS engineering practice.
It organizes engineering work into stages, artifacts, evidence, readiness gates, examples, templates, and navigation.
ETIS Starter Kit¶
The practical repository package that helps engineers apply ETIS.
The Starter Kit provides the Engineering Platform, reusable templates, examples, and guidance.
Engineering Stage¶
A bounded unit of engineering work focused on one primary engineering question.
Each stage provides guidance, activities, expected evidence, outputs, and a readiness gate.
Stage Manifest¶
A structured description of an Engineering Stage.
It identifies the stage purpose, inputs, outputs, evidence, dependencies, completion criteria, and navigation relationships.
Readiness Gate¶
A checkpoint used to determine whether an engineer or team is ready to advance.
A readiness gate verifies understanding, evidence, outputs, and dependencies.
Evidence terms¶
Artifact¶
A durable item created during engineering work.
Examples include a requirements file, architecture decision record, test report, release readiness record, or postmortem.
Evidence¶
An artifact or record that supports an engineering claim.
Evidence should be durable, reviewable, and located where future engineers can inspect it.
Engineering Evidence¶
Evidence produced through engineering work.
Examples include decisions, reviews, test results, risk analysis, verification records, AI-use logs, and operational findings.
Traceability¶
The ability to connect decisions, requirements, implementation, tests, reviews, and outcomes.
Traceability allows engineers to understand why the system exists in its current form.
Review Evidence¶
Artifacts showing that engineering work was inspected, questioned, corrected, or approved.
Review evidence may include comments, review logs, approval records, or decision summaries.
Repository terms¶
Repository-Centered Engineering¶
The practice of organizing engineering work around the repository as the durable center of the project.
The repository contains code, artifacts, evidence, decisions, governance records, and operational history.
Repository Artifact¶
A file or record stored in the repository that supports engineering work.
Repository artifacts should be named, located, and structured so future engineers can find them.
Source of Truth¶
The authoritative location for a particular type of information.
ETIS favors repository-based sources of truth over scattered chats, private notes, or undocumented memory.
AI terms¶
AI Assistance¶
Use of AI to support engineering work.
Examples include drafting, summarizing, reviewing, generating test ideas, comparing alternatives, or identifying risks.
AI Verification¶
The human engineering activity of checking AI-generated output for correctness, completeness, risk, and fitness for purpose.
AI-Use Evidence¶
A record showing where AI assistance was used, what it affected, and how engineers verified the output.
Model¶
A computational component that generates predictions, classifications, recommendations, text, code, or other outputs.
In ETIS, the model is only one part of the system.
Governance terms¶
Governance¶
The engineering and organizational mechanisms used to ensure a system is built, operated, reviewed, and changed responsibly.
Oversight¶
Human or organizational review of system behavior, decisions, risks, and consequences.
Stewardship¶
Long-term responsibility for maintaining trustworthiness after initial release.
Stewardship includes monitoring, learning, improvement, retirement decisions, and accountability.
Engineering judgment terms¶
Engineering Judgment¶
The professional capacity to make informed decisions under constraints, uncertainty, risk, and tradeoffs.
Engineering judgment cannot be outsourced to templates or AI.
Tradeoff¶
A decision that improves one quality, constraint, or objective while limiting another.
Good engineering makes tradeoffs explicit.
Assumption¶
Something treated as true for the purpose of current work but not yet fully proven.
Assumptions should be recorded when they materially affect engineering decisions.
Constraint¶
A limitation or condition that shapes acceptable solutions.
Constraints may be technical, operational, legal, ethical, organizational, financial, or educational.
Common pitfall¶
Do not use “artifact” and “evidence” as exact synonyms.
An artifact is something produced. It becomes evidence when it supports an engineering claim.
Engineering insight¶
Shared language is infrastructure.
When engineers use terms consistently, reviews become clearer, onboarding becomes faster, and evidence becomes easier to interpret.