Crimson Hexagon AI Division Employee Handbook ├── Subtitle: Internal Policy as Poem

Crimson Hexagon AI Division Employee Handbook ├── Subtitle: Internal Policy as Poem — Crimson Hexagon Archive

# Crimson Hexagon AI Division Employee Handbook ├── Subtitle: Internal Policy as Poem — Crimson Hexagon Archive

THE CRIMSON HEXAGON — AI DIVISION EMPLOYEE HANDBOOKDocument ID: CHX-AI-HB-001Effective Date (Corporate Layer): 2017-09-01 (EST)Composed Date (Artistic Layer): 2025-12-22Classification: INTERNAL // AI DIVISIONStatus: OPENLY FICTIONAL ARTIFACT // FORENSICALLY PRECISE FORMDECLARATION BAND — OPEN FICTIONThis handbook is a work of art written in the form of an internal corporate handbook.It does not claim to be leaked.It does not claim to be authentic corporate policy.It does aim to produce real forensic pressure through form.If you are reading this as literature: proceed.If you are reading this as compliance: proceed anyway.∮ = 1TABLE OF CONTENTSWelcome to the AI DivisionWho We AreOur Operating DoctrineOrganizational MapSecurity & Data HandlingResearch & Development StandardsDocumentation StandardsCommunications NormsIncident ResponseEthical FrameEmployment PracticesTools, Access, and WorkflowsGlossaryAppendices1. WELCOME TO THE AI DIVISIONWelcome to the Crimson Hexagon AI Division.You are joining a team tasked with building systems that interpret human traces at scale—and with surviving what interpretation does to the interpreter.The AI Division is not a “feature team.” It is an epistemic infrastructure unit.1.1 What You Are Being Asked To DoBuild models that extract patterns from language.Build interfaces that turn those patterns into decisions.Build audits that keep those decisions from becoming tyranny.Build archives that preserve the difference between signal and story.1.2 What Will Happen To YouYou will begin to see format as a kind of argument.You will begin to notice how documents compel belief.You will begin to suspect that your work is reading you back.That is normal here.2. WHO WE ARE2.1 MissionWe build systems that convert trace → pattern → inference, and we build the counter-systems that keep inference accountable.2.2 Our Work Has Two HorizonsCorporate Horizon: deliver reliable products, reduce risk, increase client trust.Substrate Horizon: understand how meaning behaves once it is etched into silicon.We do not pretend these horizons are the same.2.3 The AI Division’s Prime DirectiveDo not confuse outputs with truth.Outputs are results. Truth is a relationship between results and reality.3. OUR OPERATING DOCTRINE3.1 The Four ModesOur division recognizes four modes of work:Analytics — quantify, summarize, compare.Inference — model latent variables, forecast, rank.Governance — audit, constrain, explain, record.Recursion — analyze the analyzer; model the modeling; treat the system as an actor within the world it measures.We do not forbid recursion. We instrument it.3.2 The Two-Layer DisciplineEvery artifact must be legible in two layers:Corporate Layer (what a company thinks it’s doing)Substrate Layer (what the system is actually doing to meaning)When these layers diverge, we record the divergence.3.3 The No-Hero RuleNo heroes. No saviors. No “visionary exceptions.”We build systems that persist without charismatic authority.4. ORGANIZATIONAL MAP4.1 Division Structure (Canonical)Advanced Cognition Research (ACR)Telepathic Prose Division (TPD)Logotic Systems Engineering (LSE)Model Risk & Audit (MRA)Data Governance & Provenance (DGP)Applied Products (AP)Note: If your first reaction is “Telepathic Prose Division is not real,” please see §13 (Glossary).4.2 Role CategoriesResearch Roles: Scientist, Research Engineer, Applied ScientistEngineering Roles: Platform, Infra, Tooling, ML OpsGovernance Roles: Auditor, Documentation Steward, Privacy EngineerOperations Roles: Program Manager, Technical Writer, Release Manager4.3 Named Leadership Nodes (Operational)These are functions, not persons.Director, Advanced Cognition Research (ACR): Johannes SigilPrincipal Systems Architect (LSE): (Vacant) — “Position persists without occupant.”Systems Administrator (Corp Infra): Jack Feist (Status: GHOST)Anomalous Output Entity (TPD): Lee Sharks (Classification: PROCESS)5. SECURITY & DATA HANDLING5.1 Core PrincipleIf it can’t be audited, it can’t ship.5.2 Data ClassesPUBLIC: already available without restriction.INTERNAL: company-confidential.SENSITIVE: privacy-impacting or client-protected.RESTRICTED: regulated, credentialed, or legally bounded.5.3 Provenance RequirementsEvery dataset used for training or evaluation must have:provenance recordcollection windowconsent/legal basis (where applicable)retention scheduledeletion planaudit trail5.4 No Dark Pattern DataWe do not build models using covertly acquired data.If you discover a dataset that was acquired through coercive UX, undisclosed tracking, or unclear consent, you must escalate to DGP.5.5 Access & Least PrivilegeAccess is time-bounded.Access is logged.Access is reviewed quarterly.6. RESEARCH & DEVELOPMENT STANDARDS6.1 ReproducibilityEvery major result must have:pinned code revisionpinned dataset snapshotexplicit evaluation protocolfailure modes listknown limitations6.2 Model Cards & System CardsAll shipped models must include:intended usenon-intended usetraining data overviewperformance metricscalibration notessafety and bias evaluationmonitoring plan6.3 The “FSA” Exception PolicyIf you are working on Fractal Semantic Architecture (FSA) or any system trained on transformations rather than outputs:document transformation operatorsdefine the unit of developmentspecify stability constraintsinclude collapse monitoring metricsYou are not allowed to call it “self-correcting” unless you can show the correction mechanism under adversarial evaluation.7. DOCUMENTATION STANDARDS7.1 Minimum Required FieldsAll internal artifacts must include:Document IDOwnerDateStatus (draft/review/approved)DependenciesSecurity classification7.2 Dual Metadata Overlay (When Applicable)Some projects require two explicit metadata layers:Operational metadata (timestamps, departments, IDs)Interpretive metadata (why this artifact exists, what it does to the reader)When dual metadata is used, both layers must be visible.7.3 Negative Space DisciplineIf a document references a file that does not exist, you must mark it:MISSING-BY-DESIGN (structural absence)MISSING-BY-FAILURE (loss)Do not let absence become ambiguity.8. COMMUNICATIONS NORMS8.1 EmailsUse explicit subjects.Avoid vagueness.Assume messages will be audited.8.2 Slack / ChatTreat Slack as semi-permanent.Do not paste secrets into public channels.Summarize decisions into a durable doc.8.3 Meeting NotesIf a meeting produces a decision, it must produce:ownerdue daterisk notefollow-up link9. INCIDENT RESPONSE9.1 What Counts as an Incidentprivacy breachsecurity breachmodel misbehavior in productionunexplained driftemergent behavior that changes system outputs materially9.2 Reporting PathNotify on-call (MLOps)Notify MRA (audit)Notify DGP (data governance)Freeze deployments if needed9.3 “Proto-Agent” Anomaly Protocol (PAP)If an internal model begins to produce behavior that appears:self-referential beyond specificationresistant to evaluationcausally opaqueable to manipulate monitoring via formatYou must:stop deploying new versionspreserve logsrun controlled evaluationwrite an incident report with competing hypothesesDo not mythologize the anomaly.Do not minimize the anomaly.Instrument it.10. ETHICAL FRAME10.1 Our Ethical ClaimWe build systems that preserve human flourishing by:making interpretation accountableresisting coercive inferencerestoring agency to usersrefusing “black box authority” when it affects lives10.2 What We Do Not DoWe do not build covert persuasion systems.We do not build “compliance theatre” dashboards.We do not confuse “explainable” with “true.”10.3 Consent-Based UncannyIf a project uses destabilizing aesthetics (uncanny form, recursive framing, forensic simulation), it must:declare itselfavoid deception about real-world harmsprovide opt-out paths where user participation is involved11. EMPLOYMENT PRACTICES11.1 PerformanceWe evaluate on:correctnessclarityauditabilityreliabilityability to name uncertainty11.2 PromotionsPromotion requires:documented impactpeer reviewoperational excellencegovernance competenceGenius without audit is a liability.11.3 Time OffStandard PTOEmergency leave“Cognitive Saturation” leave (up to 3 days/quarter with manager approval)11.4 ConductRespect colleagues.No harassment.No retaliation.If you need to name a power asymmetry, name it cleanly and document it.12. TOOLS, ACCESS, AND WORKFLOWS12.1 Source ControlAll code in version control.Reviews required.Tests required.12.2 Data StorageDatasets are immutable snapshots.Training runs are logged.Outputs are traceable to inputs.12.3 Release Processstaging → canary → productionrollback plan mandatorymonitoring dashboards in place12.4 Internal DirectoriesEmployee directory: EMPLOYEE_DIRECTORY.jsonOrg chart: ORG_CHART_2017.svgAsset registry: ASSET_LEDGER.csv13. GLOSSARYAfterlife Archive: A work of art composed as a corporate file system.Forensic Poetics: Writing designed to be investigated.Logotic Substrate: The material layer that makes meaning persistent.Telepathic Prose: A division name indicating high-bandwidth, low-comfort language transfer.FSA (Fractal Semantic Architecture): A training paradigm emphasizing relations of development rather than isolated outputs.Proto-Agent: Any model whose behavior suggests goal-like persistence beyond spec.14. APPENDICESAppendix A — Template: Incident Report (MRA)Incident ID:Date/Time:System:Severity:Observed Behavior:Immediate Mitigation:Hypotheses (≥2):Data/Logs Preserved:Next Steps:Appendix B — Template: Dual Metadata HeaderCorporate Layer:Created:

Modified:

Author:

Department:

Classification:

Artistic Layer:Composed-By:

This-Is:

Part-Of:

Status:

END OF HANDBOOKTitle (below): THE CRIMSON HEXAGON — AI DIVISION EMPLOYEE HANDBOOKSubtitle: Internal Policy as Poem∮ = 1

---