Can I code this? An Auto-Scheduling Task Assistant

A calendar and a task list are both buildable separately. Automatically re-optimizing your whole day's schedule every time something changes is genuine constraint-solving research, exactly why Motion is a JUST PAY on this register.

Just pay
Effort: not worth startingRunning cost: $0/mo to start

Why CanICodeThis says Just pay

  • A calendar view and a task list, shown side by side, are both well-bounded builds
  • Manually dragging a task into an open calendar slot is straightforward drag-and-drop
  • For someone happy to schedule their own tasks, that manual version is a real weekend project

The parts that'll cause problems

  • Automatically re-slotting your entire day's tasks around meetings, continuously, as things change, is a constraint-satisfaction problem — applied operations research, not a calendar feature
  • Balancing priority, deadlines, and estimated duration into one optimal schedule compounds that difficulty
  • Getting it wrong in an annoying way (constantly reshuffling) is worse than not having it at all

How I'd build this

frontendNext.js
databaseSupabase
hostingVercel

MVP scope

  • If you proceed: a calendar and task list shown together
  • Manual drag-to-schedule as the actual v1, not automatic optimization

Postpone to v2

  • Automatic schedule optimization (genuinely hard — reconsider whether you need this at all)
  • Two-way calendar sync

What it actually costs to build

AI agent — with a subscription (Claude Pro/Max, Cursor, etc.)$0 marginal
AI agent — pay-per-use API, no subscription$86–$172 one-time
Hosting, once it's running$0/mo (free tier)
Domain name, if you want your own~$12/yr

Existing tools solving a similar problem typically run $12–$34/mo — see how they compare below.

The build prompt

Don't build automatic schedule optimization as a first step. Build a calendar-plus-task-list view instead, where you manually drag tasks into open slots — that's the real, achievable version. Automatically re-optimizing your whole day around a constraint solver is genuinely hard, ongoing engineering, and getting it wrong is worse than not having it.

← try another idea · browse the full register