A scheduling tool built for a real user, solving a real problem — in 24 hours.
A single-file web app that eliminates weeks of manual scheduling for competitive dance studios. No backend, no install, no spreadsheet juggling.
Every season, the Ballet Director at a competitive dance studio faces the same impossible task: scheduling 200+ groups across multiple rooms and instructors, across a full week of rehearsals, classes, and performances — entirely by hand.
The constraints are brutal:
- No dancer can be in two places at once
- No instructor can teach two groups simultaneously
- No room can be double-booked
- Standing classes must be protected as hard blocks
- Competition groups need priority placement
- The whole thing has to fit within studio hours
Getting it right takes weeks of iteration. Getting it wrong means calling dancers, reshuffling instructors, and starting over.
Build something that:
- Respects how the studio already works (CSV exports from existing tools)
- Requires zero technical setup from the user
- Generates a conflict-free schedule in seconds
- Surfaces any problems it can't solve clearly, with enough detail to fix them manually
Single HTML file, no install. The user is a dance director, not a developer. Asking her to install Node, run a server, or manage dependencies is a non-starter. The entire app — UI, logic, solver — lives in one file she can open in any browser.
CSV import, not manual entry. She already has her roster data in spreadsheets. Rebuilding it in a new format would create friction and errors. The app parses her existing files directly, with flexible column name matching so it works even if the headers aren't exact.
Conflict matrix before scheduling. Before the solver runs, users see a visual map of which groups share dancers. This surfaces the hardest constraints upfront and sets accurate expectations — if 40 groups share dancers with 30 others, no scheduler will make everyone happy.
Conflict report, not silent failure. When conflicts can't be resolved, the app doesn't hide them or crash. It names the specific dancers, groups, times, and reasons — so the director knows exactly what needs a manual decision.
Undo stack. Scheduling is iterative. The app supports full history rollback so users can experiment, see what breaks, and step back without starting over.
The scheduler runs a local constraint solver entirely in the browser — no server, no API calls. The core algorithm is a pure JavaScript bin-packing engine with backtracking:
- Builds a conflict map across all groups by identifying shared dancers
- Sorts groups by conflict density before placement (most constrained first)
- Assigns each group to a room/time slot while enforcing:
- Dancer conflicts (no dancer in two places at once)
- Instructor conflicts (no instructor double-booked)
- Room conflicts (no room double-booked)
- Standing class blocks (pre-existing classes as hard constraints)
- Custom instructor-room bans (weekend constraints, etc.)
- Backtracks and retries on constraint violations
- Sanitizes output and surfaces unresolvable conflicts with full detail
- CSV import — dancers, groups, rooms, instructors, and classes from any spreadsheet
- Conflict matrix — visual map of shared-dancer conflicts across all groups
- Free time analysis — per-room and per-instructor availability breakdown
- Instructor load balancing — view and distribute teaching hours across the schedule
- Student schedule view — per-dancer lookup across the full week
- Undo stack — full snapshot-based history with rollback
- Standing class protection — pre-existing classes block scheduling slots
- Conflict report — violations surfaced with dancer names, times, and reason
- Single file — entire app in one
.htmlfile; open in any browser, no install
Before: Weeks of manual iteration per season, high error rate, constant reshuffling.
After: Conflict-free draft schedule generated in under 60 seconds.
The app is currently in production use at a competitive dance studio, under active stress testing across a full season.
- Download
Studio_Scheduler.html - Open in any modern browser (Chrome recommended)
- Import studio data via CSV or enter manually
- Set scheduling parameters (days, hours, session length)
- Click Generate Schedule
No internet connection required after the page loads.
- Vanilla JavaScript — no frameworks
- Pure CSS — no libraries
- Local constraint solver — bin-packing with backtracking
- CSV parsing — built-in, no dependencies
Built in collaboration with Claude over approximately 24 hours. The constraint logic and UX decisions were driven by direct domain knowledge of how competitive dance studios actually operate — the fast delivery was possible because the problem was already well-understood before a line of code was written.
Part of bmadero.github.io