Steampunk uses ServiceNow to manage internal delivery and support workflows across client programs. You need to build a model that predicts whether an incident ticket will be escalated within 7 days, and the main focus of the exercise is deciding which features to engineer, why they are useful, and how to validate that they improve model performance without introducing leakage.
You are given 14 months of historical ServiceNow incident data exported at the ticket-day grain.
| Feature Group | Count | Examples |
|---|---|---|
| Ticket metadata | 10 | priority, category, assignment_group, contact_channel |
| SLA and workflow | 8 | sla_breached_before, reassignment_count, pending_reason |
| User/account context | 6 | client_tier, requester_region, contract_type |
| Temporal fields | 7 | day_of_week_opened, hour_opened, ticket_age_days |
| Text-derived fields | 5 | short_description_length, sentiment_score, keyword flags |
A strong solution should improve over a no-feature-engineering baseline and achieve PR-AUC ≥ 0.42, ROC-AUC ≥ 0.80, and precision ≥ 0.35 at 70% recall on a held-out time-based test set.