Systems · Open role


Design the structure that makes agent output usable.

An AI agent in a regulated enterprise needs more than a capable model. It needs reliable context, checks it must pass before output lands, and a record of what it did and why. You design that structure, working between domain owners and engineers.

Mission


What the role does and why it matters here.

You design the structure that makes AI agents reliable inside regulated enterprises: what context an agent can draw on, what checks must pass before its output lands, and where a person has to read. You write it down once, before the build starts, so it doesn't get relitigated under pressure. Most weeks you work alongside the founder and domain owners on live enterprise work, and the output lives in the engagement document and the codebase.

Responsibilities


What you would own.

Design the reusable context layer

For each workflow, you define the structured context an agent step can draw on, and you make that architecture call in writing so it lands in the engagement document and stays settled.

Define the validator set

With the client's compliance, platform, and engineering owners, you specify which checks must pass before an agent run lands and which are advisory. The AI Engineer builds them as code, to your spec.

Set approval boundaries

You decide where the agent stops and a person has to read, per workflow and documented before the engagement starts, so it isn't decided under pressure later.

Maintain the audit trail spec

You specify the record each agent step keeps, what it used, which validators ran, what the reviewer saw, so an auditor can reconstruct the decision from it.

Scope and phase each engagement

You write the phasing plan at the start: which workflow ships first, what it has to demonstrate before the next one begins. Not at the retrospective.

How you think and work


Six traits the work demands.

Pedigree isn't the filter. Disposition is. The six traits below are what the work actually asks of you.

  1. Agentic intuition

    You read agents the way a manager reads a direct report: when to trust the output, when to interrupt the run, when to take the wheel back.

    You've watched an automated process produce confident output that was wrong, and you intervened before it landed.

  2. Critical thinking

    Confident-sounding output gets the same scrutiny as anything else, your own work included.

    You've replaced a stakeholder's stated requirement with the better one underneath it.

  3. Curiosity

    You pull on threads. You read outside the lane. You follow a question past the first plausible answer.

    You'd rather understand why a workflow is stuck than write the design someone asked for.

  4. Agency

    You move without being told. You decide, ship, own the call. No one has to write the playbook for you.

    You've written the design that didn't exist because waiting was not going to produce one.

  5. Systems thinking, long view

    You see how the parts connect, and where this goes in three years.

    You design the model knowing the third workflow built on top of it will surface the real test.

  6. Leadership instinct

    You orchestrate work across humans, agents, and stakeholders. You switch register between a workspace ticket, an architect call, and a senior bank room in the same day without losing what you came in to say.

    You've kept a design coherent while engineers, compliance, and operators pulled in different directions.

Useful background


  • You've designed a structure (data model, taxonomy, API contract, ontology) that other teams were required to use, and defended it against pushback from outside engineering.
  • You've shipped into an environment where the cost of being wrong was real. Regulated industries count. So does anything where a bug wasn't just bad UX.
  • You've written design documents that replaced meetings: artifacts engineers built against, not retrospectives written after the fact.
  • You've sat between technical and non-technical stakeholders and kept a decision intact through both rooms.

Regulated-industry experience isn't required. Curiosity about it is.

Logistics


How the role is set up.

Engagement type
Contract or full-time. Contractors run on a defined engagement scope and can convert to full-time. Full-time runs on a yearly review with a quarterly written check-in.
Location
Remote. Lucentive is EU-based; we expect at least four hours of overlap with CET on a working day. Travel for engagement kickoffs is occasional, not weekly.
Compensation
Discussed in the written exchange, at market rate for senior people in regulated industries. The band depends on contract or full-time and on location.

Apply


We look forward to hearing from you.

Apply for Context Architect