Excel Specialist

$20–60/hrrate as listed

micro1 is contracting Excel-fluent business professionals to build realistic document tasks and judge how well an LLM produces .xlsx, .docx, and .pptx deliverables. You write the prompts, inspect the generated files, and record side-by-side preference data explaining which output a working analyst would actually accept.

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

What the work involves

The customer is training a model to create, read, and edit Office Open XML files — not just describe spreadsheets, but produce workbooks with working formulas, coherent structure, and correct formatting. Your job is to supply the hard cases. You design scenarios drawn from your own professional history (a three-statement model refresh, a headcount plan with scenario toggles, a client-ready deck built from raw exports), run them as multi-turn conversations with the model, then open the resulting files and grade them.

Most tasks are pairwise: two AI-generated solutions, one decision, and a written rationale. The rationale is the deliverable that matters. "Option B is better" is worthless; "Option B hardcodes the growth rate in row 14, breaking the sensitivity table, while Option A wires it to the assumptions tab" is the kind of note that changes model behavior. Expect to check formula logic, circular references, number formatting, named ranges, chart source data, pivot structure, and whether the output would survive a review from a VP.

What the platform screens for

Verifiable depth. Where you worked, what you built, and what broke. Micro1's screen is AI-led and follows up on specifics — naming INDEX/MATCH is not the same as explaining when you'd choose it over XLOOKUP in a shared model.
Scenario design ability. Can you invent a task that is genuinely hard for a competent model, rather than a toy exercise?
Written precision. Feedback is read by people who don't know your domain. Ambiguity is a disqualifier.
Calibrated judgment. Distinguishing cosmetic differences from substantive ones, and saying so when two outputs are equally acceptable.

Logistics

Fully remote, contractor engagement, asynchronous task queues with occasional live calls or verbal walkthroughs. Observed pay for this listing runs $20–60/hr, set by demonstrated depth and the specialization of the queue — not guaranteed, and rates on document-generation projects vary with task complexity. Volume is project-driven and can pause between phases; treat it as supplemental rather than a full schedule. No AI background required, though prior prompt-writing experience shortens the ramp.

What they look for

microsoft excel
office open xml
microsoft powerpoint
microsoft word
business analysis
document review
scenario design
analytical thinking
written communication
verbal communication

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 have at least 3 years of professional experience using Excel, PowerPoint, and Word in a business role?
Professional Office experience · required
Have you worked in a business function such as finance, operations, strategy, consulting, sales, marketing, or HR—or in an industry like banking, tech, consulting, healthcare, or retail?
Business domain background · required
Can you communicate fluently in written and spoken English?
English communication · required
Have you used generative AI tools such as ChatGPT, Claude, or similar platforms?
LLM familiarity (preferred) · preferred

0 of 4 answered

Included with this listing

Interview prep — 10 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 the candidate can articulate real-world Excel complexity (multiple tabs, formulas, data flows) and link it to business outcomes, not just technical features.
A strong answer: A strong answer names a specific business context (e.g., financial forecasting, deal analysis, budget reconciliation), describes the model architecture (data sources, calculation layers, sensitivity logic), and explains what decision or action it enabled.
Where people lose it: Candidates describe generic spreadsheet tasks (sorting data, making charts) or vague complexity ('lots of formulas') without connecting the model to a measurable business need or showing they owned the design end-to-end.