Onboarding that answers itself.
New hires ask the brain instead of interrupting the team: policies, tools, vocabulary, who-knows-what — everything your docs know, on tap in their AI.
Your AI tools don't know your business.
Week one is a hundred interruptions
Every new hire asks the same twenty questions, and a senior teammate loses a day answering them.
Policies exist but nobody finds them
The answer is in a doc, in a folder, in a drive — behind a search that nobody trusts.
Knowledge walks out the door
When someone leaves, their context leaves too. The wiki was last updated two reorgs ago.
What people & hr can do with Gyld.
Self-serve onboarding
Point new hires at the brain from day one. Policies, benefits, how-we-work docs, and company vocabulary are all askable — with the source doc linked so trust is never blind.
“How does PTO work here, and who approves it?”
Answered from your company's real data, sources attached
A glossary that maintains itself
Gyld learns your company's vocabulary from how people actually write — project codenames, acronyms, product terms — and keeps a living glossary anyone's AI can cite.
“What does "the Dagobah migration" refer to?”
Answered from your company's real data, sources attached
Who knows what
The brain builds profiles of what each person works on and owns, so "who should I ask" has an answer before the third Slack ping.
“Who owns the Stripe integration, and what are they working on right now?”
Answered from your company's real data, sources attached
Plugs into the AI they already use.
Gyld exposes the brain as an MCP server — connect it to ChatGPT, Claude, Claude Code, Cursor, or Grok in minutes. Same brain, every tool, with per-person permissions and a full request log.
You control the sources
Admins choose exactly which apps, accounts, and folders are indexed.
Permissions mirror the source
Slack, Drive, and SharePoint access carries over — fail-closed.
Sources on every answer
The brain shows its receipts, and flags observed vs inferred.
Encrypted end to end
Credentials AES-256 encrypted; your data never trains AI models.