Data Entry Operator

Aston Carter • New York, NY

22h • $45,000 - $60,000 • Remote

About The Position

Evaluate how frontier AI models handle realistic clinical work, such as creating an assessment and plan from a history and exam, medication reconciliation with interaction checks, triage decisions, discharge summaries, patient education, and clinical protocols. Assemble realistic professional files based on the work practitioners in your field actually handle. Tasks may range from compact scenarios involving a handful of files to larger scenarios that take several days to build. The goal is to create tasks that a competent professional would complete correctly but that a frontier model currently gets wrong.

This is a fully remote independent-contractor position. Work at least 10 hours per week, with no weekly maximum; full-time hours are available. Scheduling is flexible, and multi-day task timers allow work to be spread across days. Access to Claude and ChatGPT is provided through the project. Pay is listed as $40–$125 per hour USD, paid regularly through PayPal. The posting also lists an annual pay range of $45,000–$60,000 and states that no benefits are offered. The role is described as permanent and based out of New York, NY, while also identified as fully remote.

Requirements

• Active or recent clinical license: RN, NP, PA, MD or DO, PharmD, or an allied health license involving independent clinical judgment.
• At least 2 years of hands-on experience in the field preferred; candidates with less experience may be considered when they have real practical experience.
• Currently or recently performed the work as an individual contributor, rather than solely in a managerial capacity.
• Full professional or native-level written and spoken English, with strong written communication skills.
• Ability to draw on real-world clinical experience and day-to-day workflows to craft scenarios that test whether AI can perform the work.
• Comfort with ambiguity, strong attention to detail, and ability to quickly develop an accurate understanding of unfamiliar files, including material partly outside one's specialization.
• Ability to assess feedback, apply correct feedback without hand-holding, and independently look for answers when needed.
• Ability to learn unfamiliar work quickly from written materials and instructions, including when materials are incomplete.
• General familiarity with AI and LLM tools, such as Claude or ChatGPT, and the judgment to distinguish sound answers from plausible but incorrect ones.
• Baseline technology literacy, including comfort with cloud file tools such as Google Workspace, managing browser profiles, installing desktop applications, and handling files such as converting spreadsheets and creating ZIP archives.
• Availability of at least 10 hours per week; there is no weekly maximum, and consistent availability is valued.
• Must be based in the United States, Canada, or the UK.

Nice To Haves

• At least 2 years of hands-on experience in the relevant field.

Responsibilities

• Evaluate AI models' handling of clinical work, including assessment and plans, medication reconciliation and interaction checks, triage decisions, discharge summaries, patient education, and clinical protocols.
• Assemble realistic professional files and build clinical scenarios based on real-world workflows in the practitioner's field.
• Create tasks that competent professionals would complete correctly but that frontier AI models currently get wrong.
• Explain complex professional reasoning clearly and concisely, including why an AI-generated result is incorrect.
• Verify claims in documents against underlying numbers, sources, or facts.
• Interpret feedback, determine which parts are correct, and apply it independently.
• Ramp up on unfamiliar work using written materials and instructions, including potentially writing grading rubrics.

What This Job Offers

• Career Level: Mid Level
• Education Level: No Education Listed