{"id":21441,"date":"2026-10-09T10:41:03","date_gmt":"2026-10-09T17:41:03","guid":{"rendered":"https:\/\/selectoraistg.wpengine.com\/?p=21441"},"modified":"2026-10-09T10:47:42","modified_gmt":"2026-10-09T17:47:42","slug":"meet-rosetta","status":"publish","type":"post","link":"https:\/\/www.selector.ai\/ja\/blog\/meet-rosetta\/","title":{"rendered":"Meet Rosetta"},"content":{"rendered":"\n

During an incident, the answers an engineer needs are spread across tools. Telemetry sits in one place, configurations in another, and tickets in a third. Building a complete picture means knowing which tool holds which piece and how to query each one, usually under time pressure.<\/p>\n\n\n\n

Rosetta<\/a> is the conversational interface for Selector Foundry<\/a>, the agentic NetOps platform built on Selector\u2019s existing AIOps and Observability solution. You ask a question in plain language, and Rosetta works out which specialist agents can answer it, calls them in parallel, and returns one reply with the evidence attached.<\/p>\n\n\n\n

Rosetta succeeds Selector Copilot and extends it in two ways. It holds context across a full conversation, and it coordinates a crew of specialist agents behind the scenes.<\/p>\n\n\n\n

How Rosetta works<\/h2>\n\n\n\n

Rosetta draws on the context Selector\u2019s data platform has already built. The platform collects telemetry from more than 300 sources, normalizes it into one model, and correlates events across domains to establish root cause. Rosetta reasons over that correlated context, which is why its answers stay consistent and trace back to evidence.<\/p>\n\n\n\n

Every request moves through four steps:<\/p>\n\n\n\n

    \n
  1. Ask.<\/strong> You ask a question in plain language, using your own terms and acronyms. There is no query syntax to learn.<\/li>\n\n\n\n
  2. Plan.<\/strong> Rosetta reads the intent behind the question and decides what information the answer requires.<\/li>\n\n\n\n
  3. Orchestrate.<\/strong> Rosetta calls the right specialists in parallel. A correlation question goes to Loom, a ticketing question to TicketMaster, and a request for a new view to Artist.<\/li>\n\n\n\n
  4. Resolve.<\/strong> Rosetta combines the findings into one reply, with the evidence behind each claim.<\/li>\n<\/ol>\n\n\n\n
    \"\"<\/figure>\n\n\n\n

    Rosetta also works when nobody has asked yet. An incident, a schedule, an API call, or another agent can start an investigation on its own. When an engineer steps in, Rosetta is the window into that work: ask what happened, and the case, the evidence, and the next step are already there.<\/p>\n\n\n\n

    What Rosetta can do<\/h2>\n\n\n\n