Context Architect

Lucentive; United States; Remote; USD 140,000 - 210,000; Full time

Benefits offered by this job

Remote work with CET overlap

Contract or full-time engagement

Occasional travel

Job summary

Lucentive is seeking a role architect to design the operating structure for AI agents in regulated or high-consequence settings. You will translate workflow needs into architecture for agent context, validation, human review, audit records, and delivery phases, collaborating across domain and engineering teams.

Responsibilities include defining the information agents use, specifying validation checks, and establishing human review points, audit documentation, and phased engagement plans.

Qualifications

• Experience designing shared information structures (data models, taxonomies, API contracts, ontologies) and defending those choices across teams.
• Track record delivering software or systems where mistakes have meaningful consequences.
• Ability to determine when automated output is trustworthy vs. needing human intervention.
• Strong written communication guiding implementation and preserving decisions.

Responsibilities

• Define the structured information each agent workflow can use and document decisions for engineers.
• Specify required validation checks and which are advisory before outputs proceed.
• Set workflow-specific points where a human must review or approve an agent's work.
• Define audit records capturing context, checks performed, and reviewer-facing information.
• Plan engagement phases and evidence needed before expanding scope.
• Maintain clear decisions across technical and non-technical stakeholder discussions.

Skills

Data models

Taxonomies

API contracts

Ontologies

Software delivery

Trustworthy automation

Written communication

Cross-domain bridging

Job description

Role overview

Design the operating structure that helps AI agents produce dependable, reviewable work in regulated or otherwise high-consequence settings. You will translate workflow needs into written architecture for agent context, validation, human review, audit records, and delivery phases, working across domain and engineering stakeholders.

Responsibilities

• Define the structured information each agent workflow can use, and document decisions for engineers to build against.
• Specify which validation checks are required before outputs proceed and which are advisory.
• Set workflow-specific points where a human must review or approve an agent's work.
• Define audit records that capture the context used, checks performed, and information presented to reviewers.
• Plan engagement phases, including the first workflow to deliver and the evidence needed before expanding scope.
• Maintain clear decisions across technical and non-technical stakeholder discussions.

Requirements

• Experience designing shared structures such as data models, taxonomies, API contracts, or ontologies, and defending those choices across teams.
• A record of delivering software or systems where mistakes carry meaningful consequences.
• Ability to judge when automated output is trustworthy and when a process needs interruption or human control.
• Strong written communication: design documents should guide implementation and preserve decisions.
• Experience bridging domain, business, and engineering perspectives.
• Curiosity about regulated environments; direct regulated-industry experience is not required.

Benefits and work setup

• Remote work with at least four hours of overlap with Central European Time on working days.
• Contract or full-time engagement; scoped contractors may have the option to convert to full-time.
• Occasional travel for engagement kickoffs.
• Compensation is discussed based on engagement type, location, and seniority.