Production project
Production AI System-Design Casebook
Design three constrained AI systems and defend boundaries data flows failure handling evaluation and operating choices.
Project evidence
0/3 milestones · rubric 0% · pass 70%
Support workspace case
A streaming support assistant design with retrieval tools state access control and escalation.
Acceptance criteria
- Trust and authorization boundaries are visible
- Cancellation partial failure and audit paths are defined
Use: context · drivers · decision · alternatives · consequences · risks · validation · rollback · evidence
Document intelligence case
A multi-tenant document intelligence design with ingestion provenance evaluation and deletion.
Acceptance criteria
- Tenant isolation applies during ingestion retrieval and generation
- Freshness deletion and provenance flows are testable
Use: context · drivers · decision · alternatives · consequences · risks · validation · rollback · evidence
Architecture review
A comparative review that identifies bottlenecks unsafe assumptions and experiment priorities.
Acceptance criteria
- Quality latency cost and risk are evaluated together
- Recommendations identify evidence needed before scaling
Use: context · drivers · decision · alternatives · consequences · risks · validation · rollback · evidence
Criterion-level self-assessment
Project rubric
Architecture coherence
Architecture coherence for Production AI System-Design Casebook is explicit, testable, and connected to submitted evidence. · weight 20%
Current level: Missing or contradicted by the submitted evidence.
Implementation evidence
Implementation evidence for Production AI System-Design Casebook is explicit, testable, and connected to submitted evidence. · weight 15%
Current level: Missing or contradicted by the submitted evidence.
Evaluation quality
Evaluation quality for Production AI System-Design Casebook is explicit, testable, and connected to submitted evidence. · weight 20%
Current level: Missing or contradicted by the submitted evidence.
Operational readiness
Operational readiness for Production AI System-Design Casebook is explicit, testable, and connected to submitted evidence. · weight 15%
Current level: Missing or contradicted by the submitted evidence.
Security and privacy
Security and privacy for Production AI System-Design Casebook is explicit, testable, and connected to submitted evidence. · weight 20%
Current level: Missing or contradicted by the submitted evidence.
Decision communication
Decision communication for Production AI System-Design Casebook is explicit, testable, and connected to submitted evidence. · weight 10%
Current level: Missing or contradicted by the submitted evidence.
Portable evidence
Attach project records
Text, HTTPS links, repository paths, and imported Markdown/JSON/text are stored locally and included in exports.
Project review checklist
Unresolved risks
One risk per line. A passing score does not erase open risk.