[workshop-sim] Workshop Simulation Report — 2026-09-24 (Run #43000, 1000×Monte Carlo)

The issue is a simulation report identifying three specific repairs for workshop curriculum files. The first repair is to fix a regex in a shared rubric to match :white_check_mark: emoji shortcodes. The second is to add a predict-then-reveal activity to 05-agentic-workflows-intro.md. The third is to rename side-quest-07d-billing-paths.md to 07e-choose-billing-path.md and update references. Start by locating the shared rubric file and the mentioned markdown files in the repository. Verify the current regex and the emoji usage in the checklists. For the activity, examine the existing labeling exercise in 04-github-actions-intro.md as a model. Done looks like the regex matches the correct emoji format, the new activity is added, and the file is renamed with updated links.

Overview

• Date: 2026-09-24
• Students simulated: 46 × 1000 Monte Carlo runs
• Workshop steps available: 30/30
• Overall success rate: 23.6% (95% Monte Carlo interval: 23.2%–24.0%)
• Highest-dropout step: 07-first-workflow (21.2% conditional dropout among 22,507 at-risk runs; 95% Monte Carlo interval: 20.7%–21.8%)
• Lowest curriculum quality step: 04-github-actions-intro.md (overall score 5.39/10)
• Learning KPI index: 2.9/10 (active_learning 4.2 · checkpoint_quality 0.0 · scaffolding 5.0)
• Model: 2026-07-survival-model-v2 / 2026-07-assumption-model-v2 (parameter hash 2024391902)
• Limitation: synthetic results reflect explicit model assumptions; intervals exclude model and population-assumption uncertainty

Part Summary

No pages are classified as other — all 30 main steps fall inside Part 1 or Part 2.

Critical Findings

1. Content inspection reversed the top false-positive dropout. The baseline model flagged 08-run-your-workflow at 100% conditional dropout, entirely in the workflow-not-compiled category — but Step 7 (07-your-first-workflow.md) explicitly instructs gh aw compile, commit, and push with matching checklist items (lines 99–121). This was a state-tracking artifact in the simulator's step-content mapping, not a real content gap; after semantic evaluation the true bottleneck moved to 07-first-workflow (21.2%) and 05-agentic-intro (18.6%).
2. Concept-heavy Part 1 pages (04, 05, 05c, 05b) drive most real dropout, all failing on comprehension categories (concept-overload, agentic-concept-gap, agentic-classification-gap, agentic-security-gap) rather than tooling/auth friction — these are learning barriers, not access barriers.
3. Billing configuration hides inside a "Side Quest"-named page (side-quest-07d-billing-paths.md) that is functionally mandatory for Step 7 but uses the same naming convention as optional supplementary content elsewhere in the curriculum, risking learners skipping required setup.
4. Learning quality is the deeper problem, not just dropout. The cohort-wide learning KPI index is 2.9/10, driven almost entirely by checkpoint_quality scoring 0.0 across all 30 files — a rubric-detection artifact (the shared script only matches a literal "✅" character, but the workshop consistently uses the :white_check_mark: emoji shortcode), not truly missing checkpoints. active_learning (4.2/10) is a smaller but genuine gap: read-heavy concept pages should convert more passive explanation into predict-then-reveal activities.
5. The single highest-leverage repair belongs to Part 1 (lessons 00–14): fixing the rubric's checkpoint detection and adding predict/apply activities to 04-github-actions-intro.md and 05-agentic-workflows-intro.md affects every learner who reaches the core path, versus Part 2 advanced-lesson gaps which only affect the smaller subset who continue past Step 14.

Top Repairs to Prioritize

Note: some student dropout is expected and acceptable. Repairs must maintain or improve the learning KPI index — do not lower the cognitive bar or remove practice to chase headline completion numbers.

1. Fix the CHECKPOINT_RE regex in the shared rubric to also match :white_check_mark: (the emoji shortcode this workshop uses consistently), so checkpoint_quality reflects the checklists that already exist (completion impact: ↔ · learning KPI impact: ↑)
2. Add a predict-then-reveal or labeling activity to 05-agentic-workflows-intro.md (active_learning 2.4/10, lowest in the corpus) before the two-file structure explanation, mirroring the labeling exercise already in 04-github-actions-intro.md (completion impact: ↑ · learning KPI impact: ↑)
3. Rename side-quest-07d-billing-paths.md to remove the "side-quest" prefix (e.g. 07e-choose-billing-path.md) so learners do not mistake mandatory billing configuration for optional content, and make the terminal requirement explicit at the top of Step 7 (completion impact: ↑ · learning KPI impact: ↔)

Generated by 🔬 Workshop Student Simulator · copilot · auto · 247 AIC · ⌖ 9.63 AIC · ⊞ 14.8K · ◷

• expires on Sep 25, 2026, 8:00 AM UTC