中小企业 AI 高级用户 — 竞品评估

MERIDIAL
SMB AI Power User — Competitive Evaluations
Company
Meridial
Location
Remote
Engagement Type
Independent Contractor
About Meridial
Meridial places skilled, remote-first professionals into high-trust, high-precision operational opportunities for leading technology companies. Our teams work as an embedded extension of our clients' own product and operations organizations, holding the same quality bar and the same deadlines.
The Opportunity
As an SMB AI Power User on the Competitive Evaluations team, you'll help a major consumer technology company find out where its bundled AI assistant subscription falls short of the standalone assistants small businesses already pay for. You are not an AI researcher studying the problem from the outside - you are a working or experienced business owner who already runs on these tools, and your own operations are the test bed.
Your day is spent doing what you already do: drafting ad copy, planning content, answering customer messages, researching suppliers, and reconciling numbers. The difference is that you do it across competing assistants, side by side, and document precisely where each one breaks. The failures you surface are the failures real paying subscribers will hit, which is exactly why the seat calls for a genuine operator rather than an evaluator.
What You'll Do

• Run head-to-head prompts and multi-turn workflows across the client's assistant and competing paid assistants, using tasks drawn from your own business operations and marketing.

• Build and maintain a library of realistic small-business test cases spanning ad creative and copy, organic content planning, customer messaging, competitor research, pricing, scheduling, and light bookkeeping.

• Identify and document failure modes - wrong answers, unnecessary refusals, weak instruction-following, tone misses, brand-unsafe output, broken formatting, context lost across turns, unusable image or video output.

• Write clear, reproducible reports: prompt used, model and surface, expected output, actual output, severity, and why the failure matters to a paying business subscriber.

• Assess the bundled subscription on value terms - what a business owner actually receives versus what a standalone competitor subscription delivers for comparable spend.

• Stress-test assistant features inside the social and commerce surfaces where business owners meet them, not only the standalone chat product.

What We're Looking For

• Owner or operator of a small or medium business - sole proprietor, founder, or the person running day-to-day operations at a business of roughly 1–200 employees.

• Daily, habitual use of generative AI tools in that business, for both operations and marketing. Experimental or occasional use is not a fit.

• Organic presence on Instagram and Facebook for the business, maintained personally: posting, Stories, Reels, comment and message response.

• Hands-on paid advertising experience on Meta platforms - has personally built, launched, and optimized campaigns with real budget.

• Based in the United States and authorized to work here.

• Capacity sufficient to complete deliverables per the agreed scope and timeline, which may include up to 40 hours per week

• Exceptional written English - able to describe a failure clearly enough that an engineer can reproduce it without follow-up.

• Comfortable working independently against a rubric and holding a weekly output cadence.

Nice to Have

• At least one current paid AI subscription (ChatGPT Plus/Pro, Claude Pro/Max, Gemini Advanced, or similar), so competitive comparison rests on actual use.

• Runs a side business or second venture alongside a primary role - the profile that tends to test AI hardest against limited time.

• Heavy Instagram user, personally and commercially: fluent in Reels, Stories, trends, and creator tooling.

• Subscribes to two or more AI assistants simultaneously and holds an informed view on where each one wins.

• Uses AI beyond chat - automations, agents, API or no-code integrations, custom GPTs or Projects.

• Prior experience with model evaluation, red teaming, data annotation, QA, or structured user research.

• Working familiarity with adjacent SMB tooling: Shopify, Square, QuickBooks, Klaviyo, Canva, CapCut, Later, or similar.

• Industry range across the two hires - for example, one services or retail operator alongside one e-commerce or creator-economy operator.

We offer a pay range of $40-$50 per hour, with the exact rate determined after evaluating your experience, expertise, and geographic location. As a contractor you’ll supply a secure computer and high‑speed internet; company‑sponsored benefits such as health insurance and PTO do not apply.
Originally posted on Himalayas