Product Experimentation Design Reviewer

The Sage Group’s client, an AI safety and research company working to build reliable, interpretable, and steerable AI systems — with a mission to ensure AI is safe and beneficial for both customers and society — is seeking a part-time Product Experimentation Design Reviewer.

Job Description
This is a narrow role, on purpose. We’re not looking for product design broadly — we need people who excel at incremental growth product design: the unglamorous work of judging whether a change to a page or flow will move metrics. This includes thinking deeply about other potential solutions when presented with a narrow set of design directions.
Our growth team ships a high volume of experiments, and most start as a design treatment generated by Claude. Your job is to look at each one and say good or bad, and why. Your judgments will be aided by a loose set of criteria we provide, but ultimately we are looking for you to use your own uniquely human product sense and judgement. Your judgments train an automated reviewer, so consistency matters more than throughput.
Being direct about what this is: it’s evaluation and labeling work. You won’t be making comps, owning surfaces, or building anything original. Instead, you will be giving feedback on generated designs and nudging them to better and better results.

What you’d do

• Review design treatments against a written design bar and rule set, using our internal review tool

• Score and label each one, with a short written critique where the call is contested

• Flag patterns — recurring mistakes, gaps in the bar, cases the rules don’t cover

• Join a short weekly calibration with our head of design for the growth org

Who this is for

• You’ve worked directly on a growth or experimentation heavy product team. Not adjacent to one. This is the single strongest signal, well above years or seniority — incremental conversion work is a genuinely different instinct from brand, marketing or 0?1 product design

• You’ve run or analyzed A/B tests yourself.

• You reason from data and you’re comfortable being wrong in public about what will win

• You write briefly and usefully — critiques need to land for someone who didn’t see the design

• You are able to distill your product sense into clear actions. We cannot rely on vague feedback; it needs to be specific and actionable

• You’ve written or maintained design guidelines, heuristics or a review rubric
• What we’re looking for: real growth-team experience, at least one experiment described with an actual metric, and an honest reaction to the repetitive nature of the work. Portfolio or brand answers, and wins described without a number.

Duration: Through EOY plus extension potential
Compensation: $165-195hr
Location: 100% remote in USA
Hours: 2-3 hours a day, and flexible when those hours are