What the work actually involves

This is not client editing work. You are producing evidence of expert judgment that a model can learn from. Typical tasks include: recording narrated screen-capture walkthroughs of node-graph builds in Resolve's Color page; writing step-by-step task instructions that a model or a junior editor could follow to reach a specific look; grading the same clip several ways and explaining why one primary/secondary split beats another; and reviewing AI-generated editing instructions or color decisions for whether they would actually work in Resolve as described.

A large share of the volume is comparison and critique — two proposed timelines or two grades, and you write which is better and the technically grounded reason. Answers that stop at "looks better" get rejected. Answers that cite crushed blacks against a waveform reading, an incorrect color space transform in the ACES pipeline, or a Fusion node that doesn't exist in the version specified are what the pipeline pays for.

What the screen checks

  • Version and tool specificity. Reviewers probe whether you know what changed between Resolve 17, 18 and 19, and where features actually live in the UI.
  • Grading vocabulary under follow-up. Log vs. linear, CST vs. ACES, power windows and tracking, HDR vs. Rec. 709 deliverables, scopes-driven decisions rather than eyeball-only.
  • Ability to write instructions. Much of this is prose, not clicking. Clear, ordered, reproducible written steps matter as much as your reel.
  • Honest self-scoping. Editors who claim equal mastery of Fusion, Fairlight, edit, and color usually fail the follow-up on at least one.

Logistics

Remote and asynchronous, with per-batch deadlines rather than fixed shifts. Most contributors take 10–25 hours a week and can cluster it; some batches carry 24–48 hour turnarounds, so being reachable during the window matters more than working set hours. You need your own Resolve install (Studio is preferred for some codec and HDR tasks), a display you can reasonably trust, and screen-recording capability with clean audio. Pay is per-project or hourly against a scoped batch; the observed band is wide because grading and pipeline-audit tasks price well above basic annotation.