Engine 04 / Knowledge

Knowledge Engine

Turns the policies, past proposals, decisions, and tribal know-how scattered across Drive, Notion, and Slack into answers your team can actually pull up.

When this engine is worth diagnosing

Every new hire's first month is shadowing the one person who remembers how it works.
The answer exists in a doc somewhere — but it's faster to DM Sarah than to find it.
Proposals, SOWs, and policies get rewritten from scratch instead of pulled from the last good version.

How the work moves

Step 01

Gather

Bring in the docs, policies, past proposals, ticket history, and decision notes that people already DM each other to find.

Step 02

Rank

Decide which source wins when two contradict, what's stale, who can see what, and what the system must refuse to answer.

Step 03

Answer

Search, drafts, and summaries inside Slack, the help desk, or the proposal tool — every answer shows the source it came from.

Step 04

Refresh

Track unanswered questions, sources gone stale, and decisions that just got made — and feed them back in.

What ships

Source map

What lives where, who owns it, what overrides what, and what the system isn't allowed to answer from.

Agent-cited answers

Search, draft, and summary actions in the tools the team already uses — every answer linked back to the source.

Refresh cadence

A monthly review for adding new decisions, retiring stale docs, and spot-checking the answers people relied on.

Proof metrics

Use: Questions answered with a source
Quality: Answers flagged as wrong
Freshness: Sources past their refresh date

Diagnostic request

Diagnose your knowledge engine

We separate useful institutional memory from document landfill and price the first deployment. Ten thousand, fixed, credited if you continue.

Workflow map / eval harness / go-no-go