Management Consultants — Help Build Tests for AI (Ongoing, 1000+ Tasks)

We're building tests that measure whether AI systems can actually do real consulting work — and we need people who do that work for a living to design them. Most AI benchmarks for this kind of work are written by people who have never defended a market size to a client who thought it was too big, rebuilt a workplan after a scope change, or written a steerco page that had to survive a difficult room. The result is tests that look impressive and prove nothing. We're trying to fix that, and it only works if the tasks come from people who have done the job. Scope is strategy and transformation consulting: engagement management, research and market intelligence, PMO, and commercial due diligence. Firm experience — MBB, Big 4 advisory, boutique, or in-house strategy — rather than general freelance business consulting. What you'd actually do: - Take a piece of work you've genuinely done — a market sizing build, a competitor benchmarking table, a steering committee pack, a hypothesis tree, a CDD findings section — and turn it into a task an AI system has to attempt - Define what "correct" looks like, and just as importantly, what a plausible-looking wrong answer looks like - Review tasks other consultants have written and flag anything that wouldn't survive a partner review or a client steering committee The interesting part: you'll see exactly where these systems fall over on real work in your field. Most people guess about that. You'd know. This is not a one-off. We have over 10,000 tasks to work through and we're bringing on up to 100 practitioners to do it. Individual tasks pay between $50 and $250 depending on complexity, and there is steady ongoing work for people whose tasks hold up well. Who we're looking for: - Engagement managers and project leaders — workstream status roll-ups, hypothesis trees and workplan builds, steering committee packs, scope change assessment - Research analysts and market intelligence — market sizing builds, competitor benchmarking tables, expert interview synthesis, data room summaries - Transformation PMO leads — milestone and dependency tracking, benefits realisation tracking - Commercial due diligence analysts — CDD findings, cost baseline and savings opportunity models - Consulting firm or in-house strategy experience, any level from analyst to principal - 10+ hours a week you can spend on this Important: you'd author from representative examples you construct yourself. Never real client deliverables, data room contents, expert call notes, or confidential findings. Paid per task completed. We work with 100+ practitioners, so everything runs async and in writing — no meetings. To apply, answer the questions below. We read every answer — the question about junior mistakes matters more to us than your years of experience.