Medical Coder

$25–40/hrrate as listed

Certified professional coders review de-identified medical records and assign or verify ICD-10, CPT, and E&M codes to build training and evaluation datasets for AI systems. Work is remote contract project work through micro1, billed hourly at an observed $55–68/hr.

Summary, eligibility guidance and screening preparation are prepared by Sidequest from the platform's posting. Check the source listing before applying.

The work

You will read clinical documentation — office notes, operative reports, discharge summaries, ED encounters — and either assign codes from scratch or audit codes an AI model produced. Most tasks resolve to a defensible answer plus a written rationale: which guideline, which index entry, which excludes note, which MDM element drove the E&M level. Expect a steady mix of straightforward encounters and deliberately hard edge cases where documentation is incomplete, contradictory, or supports more than one defensible code.

A meaningful share of the job is disagreement resolution. When your code differs from a model output or another coder's, you will be asked to explain the gap in a way a non-coder engineer can follow. Coders who can cite the actual guideline language, rather than saying "that's how we've always done it," are the ones who last on these projects.

What the screen looks for

An active AAPC CPC (or equivalent credential) — this is a hard gate, and micro1 verifies it.
Guideline fluency under follow-up. Screeners probe past the first answer: why not the other code, what would change your mind, which convention applies.
E&M leveling depth, specifically under the 2021+ MDM and time-based framework, not the old bullet-count method.
Written clarity. Rationales are the deliverable as much as the codes are.
Access to current ICD-10-CM/PCS and CPT references, plus official E&M guidelines.

Logistics

Fully remote, contractor engagement, no AI or annotation background required. Work is largely asynchronous through a web annotation tool, with occasional live calls for calibration or guideline clarification. Volume fluctuates with client demand — treat it as part-time and variable rather than a steady 40 hours, and expect an unpaid calibration or trial batch before full onboarding. Pay of $55–68/hr reflects rates observed on this listing and can vary by project, specialty, and task type.

What they look for

Medical Coding
ICD-10
CPT
CPC Certification
Evaluation & Management

Before you apply

Would you pass this screen?

Drawn from what this listing actually requires. Answered in your browser — nothing is sent to us, and we do not store your answers.

Do you currently hold an active Certified Professional Coder (CPC) certification from the AAPC that has not expired?
Active AAPC CPC Certification · required
Do you have access to current-year official ICD-10 and CPT coding manuals, as well as up-to-date E&M guidelines?
ICD-10 and CPT Coding Books · required
Do you have at least two years of professional experience coding medical records across multiple clinical documentation types (e.g., inpatient, outpatient, surgical, E&M visits)?
Clinical Documentation Experience · preferred
Have you previously used digital annotation tools, participated in data labeling projects, or worked with data-driven platforms in a healthcare or technical setting?
Digital Annotation Tool Familiarity · preferred

0 of 4 answered

Included with this listing

Interview prep — 11 questions

Matched to what this platform's screening actually measures. Answer each out loud in 60–90 seconds: claim → concrete example → implication.

What it tests: Whether you follow the coding standard of coding only what is documented, and how you handle the tension between clinical reasoning and coding rules when documentation is incomplete.
A strong answer: You explain that you code to the documentation as written, not what you infer clinically; if severity or laterality is needed for correct coding but absent, you acknowledge that gap and either query the provider or flag it rather than assume, because assumptions introduce errors into training data.
Where people lose it: Saying you'd 'fill in the blanks' based on clinical sense, or that you'd code what 'should have been documented' — that corrupts the dataset and signals you don't respect the boundary between clinical judgment and coding rules.