Personalized Life Assistant Expert

What you'll do

This is evaluation work built around your own habits. A leading AI research organization wants to know whether its assistants genuinely help with personal, high-context tasks, and it is asking people who already lean on AI tools in daily life to be the judges.

In practice you look at AI responses to personal requests and decide whether they are useful, tailored to the person, realistic and actually successful at the task. The ad names the areas covered: food, health, productivity, careers and learning. The main skill is explaining why an answer falls short: generic advice that ignores stated constraints, a plan that would not survive a real week, a recommendation that is unsafe, or a response that simply stops halfway.

The ad does not describe the task format (single responses, side-by-side comparisons or multi-turn conversations), so expect to learn that during onboarding.

Who fits

• Heavy personal use of LLM products, not just occasional queries
• Experience using AI for multi-step work: planning, research, decisions or personal workflows
• Familiarity with tools such as ChatGPT, Claude, Gemini, Perplexity, Cursor, Windsurf, Codex or other AI agents
• The ability to say clearly what makes an output good, bad, incomplete, unsafe or unrealistic
• Strong written judgment and attention to detail

No degree, profession or years of experience are required. That makes this one of the more open Mercor listings, since the qualification is how you use AI. It also means your written rationales carry the whole application, so write them the way you would want an evaluator to read them.

What it pays

$50–200 per hour, Mercor's published range. The band is unusually wide (a 4x spread) and the ad does not say what places you within it. With no credentials required, plan around the lower end until you see your offer.

Expected commitment is 20–40 hours a week, which is closer to a part-time or full-time job than a side gig. Payments are weekly on Stripe or Wise.

Worth knowing

Good:

• No formal credentials required; everyday AI fluency is the qualification
• A ceiling of $200/hour is rare for non-specialist evaluation work
• Weekly payment on Stripe or Wise, published in the ad
• Mercor showed one hire on this listing this month, so it is active
• Fully remote and on your own schedule

Less good:

• US residents only
• The pay band is very wide with no published basis for placement
• 20–40 hours a week is a large commitment for evaluation work
• H-1B and STEM OPT holders are excluded
• Independent contractor: projects can be extended, shortened or ended early

About this listing

Posted on Mercor as an hourly remote contract restricted to the United States, confirmed open on 24 September 2026. Pay, hours and the requirements above are the ad's own; the client is described only as a leading AI research organization. See Mercor.