Design Jira bug-to-team classification system
Design Jira Bug-to-Team Classification System
Design a system that automatically classifies incoming Jira bug tickets into the most appropriate owning team. The system should also produce a report for customer companies detailing bug trends.
Assumptions:
- Each ticket contains fields such as title, description, component/service, logs/stack traces, priority, and customer/company.
- Historical ticket assignments made by humans can be used as training labels.
- The system should suggest or auto-route tickets to teams, with the ability for human overrides.
Deliverables:
- Define system requirements and success metrics.
- Outline data sources and the labeling strategy.
- Propose a model approach (from baseline to advanced) and feature design.
- Design the online serving architecture, including routing and a human-in-the-loop mechanism.
- Describe the training pipeline, evaluation process, and monitoring for model drift.
- Design the reporting layer for customer-facing analytics.
AI interview prep
Core dilemma
Balancing the accuracy and latency of an ML model for real-time ticket routing against the complexity of training, deployment, and ongoing maintenance, while also enabling effective human oversight and reporting.
3 tradeoffs and 3 stress tests locked
Interview prep
Unlock the interviewer X-Ray rubric, tradeoff matrix, stress tests, clarifying questions, and key concepts.
Design Jira bug-to-team classification system
Design Jira Bug-to-Team Classification System
Design a system that automatically classifies incoming Jira bug tickets into the most appropriate owning team. The system should also produce a report for customer companies detailing bug trends.
Assumptions:
- Each ticket contains fields such as title, description, component/service, logs/stack traces, priority, and customer/company.
- Historical ticket assignments made by humans can be used as training labels.
- The system should suggest or auto-route tickets to teams, with the ability for human overrides.
Deliverables:
- Define system requirements and success metrics.
- Outline data sources and the labeling strategy.
- Propose a model approach (from baseline to advanced) and feature design.
- Design the online serving architecture, including routing and a human-in-the-loop mechanism.
- Describe the training pipeline, evaluation process, and monitoring for model drift.
- Design the reporting layer for customer-facing analytics.