What the work actually is
You write project scenarios the way they arrive in real life: an ambiguous kickoff brief, a vendor slipping two weeks before a hard external date, a sponsor who changed the success criteria without saying so. Each scenario ships with a reference answer that shows the planning reasoning — why this dependency gets crashed and that one gets descoped, which stakeholder hears about the slip first and in what form, what the risk register should have caught. Then you review model output against it, scoring for realism, completeness and stakeholder awareness, and writing up why a given plan fails.
The hardest and most valued part is the last responsibility on the list: flagging plans that read polished but would fail in delivery. Models produce confident Gantt-flavoured prose, generic RAID logs and mitigation columns that say "monitor closely." Your job is to name the specific way it breaks — the critical path that ignores the client's approval cycle, the resource plan that double-books the one person who can sign off, the comms plan that escalates to a steering committee that would never convene in time.
What the screen looks for
AfterQuery's screening is AI-led and follow-up heavy. It probes for named projects with real scope, budget or stakeholder counts, and for delivery decisions you personally owned rather than observed. Expect it to push on tradeoffs — what you cut, who objected, what it cost — and to test whether you can articulate the reasoning behind a plan, not just the plan. Certifications (PMP, PRINCE2) and agile/waterfall breadth help but do not substitute for lifecycle ownership.
Logistics
- Remote, fully async — no standups, no client calls, no fixed hours
- Minimum 10 hours per week; you set the volume above that
- Paid weekly via Stripe
- Ongoing task flow rather than a fixed-term engagement
- Heavy written output, so writing speed and clarity directly affect your effective rate