Test design and production decisions against concrete cases. Start with the decision you need to make, then use the linked material to build a checkable result.
Learning path
Connect the path
The foundations define the task and trust boundary. Later modules test whether a result should be used, revised, or sent to a person.
Advanced decision quiz
Test whether the same task contract survives changed wording, evidence versions, and model settings.
Evidence decision quiz
Find the failing layer before accepting a generated answer or effect.
Applied decision quiz
Test whether the generated artifact is safe and sufficiently supported for its destination.
Evaluation code lab quiz
- Prompt evaluation code labs
- Code lab: reject unknown evidence IDs
- Code lab: score answered cases and abstentions
- Code lab: find pairwise judge order changes
- Code lab: block a critical prompt regression
Agent workflow quiz
Apply the boundaries to short operational decisions.
Reasoning checks quiz
Interpret what each reasoning pattern proves and what remains unverified.
Prompt foundations quiz
Practice the decisions that make a basic prompt testable.
Evaluation governance quiz
Practice the decisions that make an evaluation trustworthy.
Multimodal evidence quiz
Choose what each source can support and when review is needed.
Safety decisions quiz
Choose the correct boundary for common support requests.
Reasoning patterns quiz
Test stop rules, verification gates, and cost limits.
Software delivery quiz
Test the boundaries between model suggestions and verified release decisions.
Data workflows quiz
Test cohort, label, numerical, and chart boundaries.
Multilingual workflow quiz
Test integrity, authority, and parity boundaries.
Accessible output decisions
Test image, media, structure, wording, and release boundaries.
Browser-agent decisions
Check observation, target identity, approval, and receipt reasoning.
Real-time voice decisions
Check turn, playback, entity, and outcome boundaries.
Synthetic evaluation decisions
Test validity, coverage, disclosure, and split boundaries.
Generated media decisions
Test audio, pronunciation, shots, and final assembly boundaries.
- Generated speech and video prompt decisions
- Project: release an audio and video guide for a fictional exhibit
Spreadsheet workflow decisions
Test formulas, edit scope, export safety, and release evidence.
Specialist coordination decisions
Test agent scope, merge conflicts, and lifecycle limits.
Research synthesis decisions
Test selection, event rates, design differences, and release wording.
Tutoring decisions
Test help stages, evidence-based feedback, rubric review, and privacy.
UI-to-code decisions
Test visual evidence boundaries, behavior, reuse, and release checks.
Document delivery decisions
Test counts, structure, caveats, and actual export checks.
Meeting-record decisions
Test attribution, approval scope, action ownership, and release.
Architecture decision checks
Test planning assumptions, arithmetic, contracts, and rollback.
API documentation decisions
Test public fields, errors, examples, compatibility, and release state.
Infrastructure change decisions
Test plan actions, state boundaries, receipts, and postchecks.
Incident communication decisions
Test denominators, hypotheses, action receipts, and release authority.
Authorized retrieval decisions
Test access, ranking, version, and claim-support boundaries.
Memory lifecycle decisions
Test write authority, namespace scope, correction, forget, and isolation.
Recurring workflow decisions
Test the failure paths in a repeated AI task.
Terminal-agent decisions
Test command evidence, permissions, patch scope, and verification.
Chart-creation decisions
Test marks, gaps, rates, lineage, and accessibility.
Email and calendar decisions
Test quoted claims, time zones, approval, and effect receipts.
Location-aware decisions
Test address, coordinate, travel, timing, and receipt boundaries.
Interview analysis decisions
Test evidence and interpretation boundaries.
Forecast decisions
Test temporal evidence and inventory boundaries.
Security alert decisions
Test evidence and response boundaries.
Survey decisions
Test question and denominator boundaries.
Support case decisions
Test account and communication boundaries.
Accessibility review decisions
Test the boundaries of the packet.
Product analytics decisions
Test the boundaries of the packet.
Localization QA decisions
Test the boundaries of the packet.
Product experiments decisions
Test the boundaries of the packet.
Invoice exception review decisions
Test the packet's decision gates.
Database backfill decisions
Test the packet's decision gates.
Search relevance evaluation decisions
Test the packet's decision gates.
Annotation quality workflow decisions
Test the packet's decision gates.