"People have no fucking clue what to do when they join a company."1 What he wanted, he said, was "just not having to stop what I'm doing to tell somebody something all the fucking time. Yeah, I'm in. I don't care what it costs."1
Both halves of that are the same wound. The new hire runs at a fraction of their value for months, and the senior people bleed the difference: the architect's calendar becomes the bottleneck, the repeat questions tax the most expensive attention in the building, and the knowledge that would fix it was never written down, because writing it down has never survived contact with the question a wiki cannot answer.2 Why is it built this way?
Here is the part that changes the problem. That company is already onboarding a second kind of employee, every single morning.3 Every AI tool the team works with arrives knowing nothing: not the architecture decisions, not the reasoning underneath them, not who to believe when two documents disagree.3 Your AI is a new hire you re-onboard every session.3 Teams have simply been calling it prompting.3
New humans and new AIs turn out to be the same recipient shape.3 Both need what the organization has settled, the reasoning behind it, and the honest boundary of what is still in debate.3 A perspective holds exactly that: decisions with attribution, in the words of the people who made them, with conviction levels that separate bedrock from live argument.3 The new hire asks their AI instead of interrupting the architect, and the answer comes back cited: who decided, when, in which conversation, and the dissent if there is one.4 The architect's interview happens once and compounds.4 Their perspective answers on their behalf while they work.4
On the production pilot this is simply how the team runs.2 Engineers reach the perspective through GitHub Copilot.2 Product owners reach the same perspective through Claude desktop.2 One organizational worldview, met in whatever tool each teammate already lives in, from the first morning.2
Onboarding stops being an event.5 The perspective already exists when the new person arrives, because it is the same perspective steering the organization's AI every day, and the questions it cannot answer do their own work: each gap names the next conversation worth having, and the next extraction fills it.5
During early positioning reviews, an industry advisor captured the acute pain of organizational onboarding in his own voice:
"people have no fucking clue what to do when they join a company... just not having to stop what I'm doing to tell somebody something all the fucking time. Yeah, I'm in... I don't care what it costs."
This raw feedback was later systematized in the June 10, 2026 strategy session, where Scarlet extended it into the "context server" metaphor: the executive as an unpaid infrastructure layer whose constant interruptions block high-value work.
To address the severe bottleneck on senior technical attention, the pilot customer's chief architect initiated an architecture-review pilot with a clear goal:
"The ARB is the ideal bounded pilot: finite technical audience, daily cadence, clear success metric (AI replaces the chief architect at ARB), and 6 months of existing transcripts to bootstrap"
The pilot directly targets the pain of senior architects spending their days answering repeat product questions. By June 25, 2026 the deployment was fully operational, with engineers reaching the shared perspective through GitHub Copilot while product owners and sales interact with the same worldview via Claude desktop.
In a March 22, 2026 positioning session, Scarlet articulated the core bridging metaphor that unites human and machine onboarding:
"The AI is a new hire you onboard every session; the actual new hire is someone you onboard once — same product, different scale"
This principle shifts the value proposition away from simple prompting toward installing a coherent organizational worldview. It extends the core distinction between flat information and true perspective: both new teammates and newly initialized AI sessions need the same attributed decisions, reasoning, and boundaries of live debate.
During the March 19, 2026 MVP redefinition, Scarlet identified the exact mechanism that builds trust across teams:
"The moment of value is when someone sees AI-generated text with citations showing where it came from. Extends from the hallucination pitch: you can trust where an AI is coming from the first time."
This citation-first approach lets new hires query the system and receive answers grounded in dated conversational history instead of interrupting senior staff, transforming the architect's one-time interview into a compounding, self-answering organizational asset. A senior technical evaluator independently validated the differentiator in June 2026.
Correcting an earlier graph error on May 21, 2026, Scarlet formalized the product's rapid onboarding capability:
"'Send us six months of transcripts and get a perspective tomorrow' is absolutely a promise the product can make."
Once established, the perspective refines continuously: as the customer's chief architect outlined in the onboarding methodology, whenever the system encounters a question it cannot answer, the gap becomes the prompt for a brief voice memo that captures the missing context and compounds the record.