How Edge unblocked 130 users in one night

The problem they were solving

Every Marker.io ticket carried real detail. However, it's exactly what makes 100+ of them unreadable. A product director can't sit down and absorb hundreds of individual reports, each with their own context, and come out the other side with a clear plan. The information was there. The capacity to process it wasn't.

‍

What they built

Shahraiz Tabassam, Director of Product, connected Marker.io's MCP server to Claude Code. The flow looked like this:

• Shahraiz gave Claude the context it needed and pointed it at the full ticket backlog.
• Claude queried the MCP server and pulled tickets, screenshots, and network data the same way a developer would.
• The output came back as a structured report: executive summary, criticality ratings, issues grouped by category, duplicates flagged, repeat reporters called out, technical patterns surfaced.
• A second pass produced a deduplicated list of user emails needing manual database updates.
• A third pass ran across non-assessment issues and surfaced blind-spot problems the team didn't know existed.

‍

Where they are now

The release shipped on the back of that first report and unblocked roughly 130 users in a single night. The database cleanup got done without developers digging through tickets manually. The blind-spot issues, including video recording problems affecting seven users and phone validation errors affecting three, got added to the roadmap before they grew. Shahraiz is now exploring the same setup for auto-replying to tickets about already-fixed issues.

‍

Why this is worth stealing

‍

• The MCP server lets a product team turn raw ticket volume into a narrative without involving engineering.
• The same pattern handles triage, deduplication, and blind-spot detection, all from one connection.
• The bottleneck shifts from reading tickets to deciding which ones to act on.

‍