Independent portfolio project
Media QC Guidance Copilot
Built from years of hands-on streaming QC and metadata operations, this design study explores how an AI assistant could find relevant guidance and show the evidence behind its answer.
Scope: independent portfolio work. This is a design standard, not a claim of employer deployment, approval, adoption, or measured performance.
Problem and goal
Media-quality guidance can be dense and context-dependent. Under time pressure, a fast, plausible answer is dangerous when an operator cannot see what supports it—or when the available evidence is too weak to support an answer at all. The goal is to find relevant guidance faster while keeping uncertainty visible and leaving the final decision with the operator.
Proposed workflow
- Operator question
- Find approved guidance
- Assess the evidence
- Answer with sources—or stop
- Operator decides
Proposed safeguards
- Search approved guidance first. Build the response only from sources the system is permitted to use.
- Show the support. Return the answer with its sources and clearly state any uncertainty.
- Stop when evidence is weak. Do not fill gaps with a plausible-sounding answer.
- Leave the decision with the operator. The person reviews the evidence and owns the consequential decision.
This specification draws on experience with video, multitrack audio, timed text, metadata, regional context, severity, and escalation. It has not been implemented or evaluated as a production system.
No employer system, data, screenshot, workflow, client detail, performance metric, or production result is represented here.