Hiring nowAlignment and scalable oversight

Model Behaviour and Character Designer

Also called Model Behaviour / Character Designer · Model Personality Lead · Model Spec Writer

A Model Behaviour and Character Designer defines and tunes how a model reasons, communicates and handles values trade-offs. They make the product's behaviour deliberate, consistent and defensible, instead of an accident of training.

What they do

The work starts with writing down how the model should behave: tone, honesty, refusals, how it weighs competing values, how it handles difficult users. Then the designer turns that into something testable. They draft specification text, build evaluation sets, and design preference comparisons. They read large numbers of model outputs, find where behaviour drifts from intent, and work with post-training researchers to adjust data, prompts or training. They also write clear rationales so others can follow the reasoning. Much of the week is close reading, careful writing and small experiments.

What a strong hire delivers

In the first months, a strong hire delivers a written behaviour specification that engineers and researchers actually use. They set up evaluations that track the traits the company cares about. They name the main gaps between intended and observed behaviour, and they run at least one training or prompting change that measurably closes a gap. Disagreements about values get decided and recorded.

How you can tell they are good

Ask them to take a behaviour they care about, such as honest disagreement, and show how they would specify it, test it and tune it. Strong people move easily between principle and example. They show their own failed attempts. They can explain why a trait measured well but felt wrong, and what they changed in response.

Why it is hard to fill

Hardly anyone has done this before, and titles vary widely, so searching by title finds very little. The work needs humanities-level judgement and enough machine learning literacy to run real experiments. Most people have one half and are keen to learn the other. Labs and consumer AI companies compete for the few who have both, so expect a thin field.

Where to find them

Look at researchers in preference learning, technical safety and alignment who write well and care about values. Also look at people from philosophy, linguistics, writing or product design who have moved into post-training work. They sit in frontier labs, consumer AI companies and safety organisations, often under titles like research scientist, member of technical staff or policy and behaviour lead.

Indicative pay

Not enough published evidence yet to give a fair range for this role, which is common for work this new. We benchmark it for you as part of a search.

Interview questions for a Model Behaviour and Character Designer

  1. Describe a model behaviour you specified, tested and changed. What was the intended behaviour and how did you know it improved?

    Strong answers give a concrete trait, the written intent, the evaluation used and the evidence of change. Look for honest account of side effects. Worry if they only describe opinions about good behaviour with no measurement.

  2. How would you write guidance for when a model should decline a request versus help with caveats?

    Good answers set out the competing values, give worked examples at the boundary and explain how to handle ambiguity. They accept some errors either way. Worry about rigid rule lists or a single value overriding all others.

4 more questions, with what a strong answer covers, are in the Role Kit.

Job description you can post

We are looking for someone to define and shape how our model reasons, communicates and handles difficult trade-offs. You will turn broad values into clear written guidance, tests and training changes, working closely with research and product. The aim is a model whose behaviour is deliberate and consistent.

What you will do

  • Write and maintain a clear specification of intended model behaviour
  • Design evaluations and preference comparisons for traits such as honesty and tone
  • Review model outputs to find gaps between intent and behaviour
  • Work with post-training researchers to change data, prompts or training
  • Document the reasoning behind values decisions
  • Advise product teams on behaviour in edge cases

What you will bring

  • Demonstrated work specifying or evaluating model behaviour
  • Working knowledge of preference learning and post-training methods
  • Experience designing experiments and reading results critically
  • Excellent writing, shown in samples we can read
  • Sound judgement on ethical and values questions
  • Ability to disagree constructively with researchers and product leads

Nice to have

  • Background in philosophy, linguistics or related humanities
  • Published or shared alignment research
  • Experience with consumer-facing AI products

Free to use and adapt. If it saves you time, a link back to this page is appreciated.

Skills that matter

  • Alignment researchWorkingStudy methods to make behaviour meet intended objectives
  • Preference learningWorkingLearn behaviour from comparisons and reward signals
  • Model evaluationWorkingMeasure capability and failure modes credibly
  • Experimental designWorkingChoose comparisons that identify credible effects

Other roles in alignment and scalable oversight

Hiring nowScalable Oversight ResearcherA scalable oversight researcher designs and tests ways to supervise AI systems on tasks too hard for a person to judge easily. The work shows…EmergingReward Hacking ResearcherA Reward Hacking Researcher finds and studies cases where models exploit the reward or evaluation signal instead of doing the intended task, then…Hiring nowModel Behaviour ResearcherA Model Behaviour Researcher studies how a model's behaviour shifts across contexts and training, and shows what caused the shift. That gives the…

Related roles in other families

Hiring nowPost-Training Research EngineerPost-training and preference optimisationA Post-Training Research Engineer adapts pretrained models into ones that behave as the company needs. They turn preference data and reward signals…Hiring nowReward Modelling ResearcherPost-training and preference optimisationA reward modelling researcher builds models that learn what good output looks like from preference comparisons. They give post-training teams a…Hiring nowReinforcement Learning Research EngineerPost-training and preference optimisationA Reinforcement Learning Research Engineer improves how a model behaves by training it against interactive signals. They matter because reward gains…Hiring nowModel Evaluation EngineerCapability evaluation and benchmarkingA Model Evaluation Engineer builds the tests that show what an AI system can do and where it breaks. Without them, a team cannot tell whether a…

How we wrote this

Role definitions are our own editorial work, checked against occupational classifications, standards, research and employer pages. Titles vary by company, and this is a description of the work, not a job advert.

← Alignment and scalable oversightAll new job families

Start a search

Hiring a Model Behaviour and Character Designer?

Tell us the work. We find the people already doing it, whatever their job title says, and you meet them already interviewed for your role.