The work

You will be assigned to one of two tracks. Question authoring means designing original problems in your engineering subdomain that require real reasoning rather than lookup: a self-contained problem statement, a difficulty rating (Medium for intro undergraduate, Hard for advanced undergraduate, Expert for post-graduate and above), one defensible correct answer, nine distractors that a competent expert could plausibly arrive at through a specific misstep, a markdown Chain-of-Thought solution with clean intermediate steps, and one to five references from peer-reviewed journals, standards bodies, or university repositories. Question verification means auditing pre-written items for ambiguity, missing givens, unit or sign errors, unsolvability, and distractors that are wrong for the wrong reasons — then editing, re-rating difficulty, and documenting what you changed and why.

Covered domains include semiconductor design and manufacturing (VLSI), control science and engineering, mechatronics, reactor and plant design and separations, reservoir engineering and maintenance, and bioinstrumentation and biotechnology. You are expected to work inside your actual subdomain, not adjacent to it; graders notice when a nuclear item is written by someone whose depth is in process control.

What the platform screens for

Mercor's screening is AI-led and follows up hard on specifics. Expect to be asked what you personally derived, measured, or fabricated — not what your lab worked on. The two things that most often decide the outcome: whether you can produce a genuinely expert-level item on demand (the distractor set is the tell — nine random wrong numbers reads as low effort, nine wrong numbers each traceable to a named error does not), and whether your difficulty calibration matches a shared rubric rather than your personal sense of "hard." Written English is assessed directly, since the deliverable is prose.

Logistics

  • Fully asynchronous and remote; no scheduled calls or shift windows.
  • Expected commitment of 10+ hours per week, with throughput and quality scores affecting continued assignment.
  • Pay of $61–77/hr reflects observed rates for this listing and varies by domain scarcity, task type, and review outcomes — it is not a guarantee.
  • Credentials are verified, and work is typically covered by NDA; benchmark items you author are not yours to reuse or publish.