# Anubis Labs

> Governed AI systems for regulated workflows: secure agent runtimes, enterprise knowledge platforms, compliance intelligence, and evidence-backed automation.

Canonical site: https://anubislabs.ca

Anubis Labs at anubislabs.ca is a Calgary-based AI systems lab. It is distinct from other organizations named Anubis Labs in Web3, subsurface imaging, malware reporting, and bot-protection tooling.

## Core Pages

- [Anubis Labs | Governed AI Systems for Regulated Workflows](https://anubislabs.ca/): Anubis Labs builds governed AI systems, secure agent runtimes, enterprise knowledge platforms, and compliance intelligence with evidence, approvals, and forensic traceability.
- [AI Solutions](https://anubislabs.ca/solutions): Governed AI solution patterns for policy-gated agent runtime, enterprise knowledge retrieval, compliance intelligence, and secure operational automation.
- [AI Architecture and Delivery Services](https://anubislabs.ca/services): Pilot, production hardening, security review, evaluation planning, rollout support, and audit-ready delivery for governed AI systems.
- [Security and Governance](https://anubislabs.ca/security): Security model for governed AI systems: identity, access control, data handling, auditability, sandboxing, egress controls, and deployment flexibility.
- [AI Case Studies](https://anubislabs.ca/case-studies): Source-backed case studies covering pursuit intelligence, staff intelligence, project experience retrieval, governed knowledge assistants, document review, and offline AI systems.
- [Delivery Process](https://anubislabs.ca/process): Delivery process from discovery and governance through prototype, evaluation, control hardening, deployment, and audit.
- [About Anubis Labs](https://anubislabs.ca/about): Calgary-based AI systems lab focused on governed AI systems for operational teams and regulated environments.
- [AI Discovery](https://anubislabs.ca/ai): Canonical Anubis Labs Inc. identity, disambiguation, service boundaries, and machine-readable discovery links for search engines, AI agents, and answer systems.
- [Resources](https://anubislabs.ca/resources): Practical Anubis Labs resources for governed AI architecture, enterprise retrieval, compliance intelligence, security posture, and delivery planning.
- [FAQ](https://anubislabs.ca/faq): Answers about Anubis Labs services, governed AI systems, deployment models, privacy posture, architecture reviews, and pilot scoping.
- [Contact Anubis Labs](https://anubislabs.ca/contact): Request an AI architecture review, governed AI pilot, security review, or delivery conversation with Anubis Labs.

## Solutions

- [Policy-Gated Agent Runtime](https://anubislabs.ca/solutions/agent-runtime): Secure agent runtime architecture with policy gates, approvals, budgets, sandboxed tool execution, trace logs, and deployment controls.
- [Enterprise Knowledge Platform](https://anubislabs.ca/solutions/knowledge-platform): Governed retrieval, semantic search, citation chains, RBAC-aware knowledge access, and evidence-backed answers for enterprise teams.
- [Compliance Intelligence](https://anubislabs.ca/solutions/compliance-intelligence): Compliance intelligence systems that combine deterministic checks, evidence capture, risk scoring, jurisdiction awareness, and human review.

## Resources

- [Governed AI Systems](https://anubislabs.ca/resources/governed-ai-systems): A governed AI system combines model behavior with access control, evidence retrieval, policy checks, approval workflows, monitoring, and operational ownership.
- [Enterprise RAG Governance](https://anubislabs.ca/resources/enterprise-rag-governance): Enterprise retrieval needs source authority, permission-aware search, citation quality, document freshness, and refusal behavior when evidence is weak.
- [AI Agent Security](https://anubislabs.ca/resources/ai-agent-security): AI agents need constrained tools, least-privilege credentials, approval gates, egress controls, budgets, and replayable run logs before they touch production workflows.
- [AI Compliance Checklist](https://anubislabs.ca/resources/ai-compliance-checklist): AI compliance work should combine deterministic rules, evidence capture, human review, risk scoring, and a record of why each decision was made.
- [Offline and VPC AI Deployment](https://anubislabs.ca/resources/offline-ai-deployment): Offline, local-first, and VPC AI deployments trade convenience for stronger data control, predictable network boundaries, and tighter operational responsibility.
- [What Is Governed AI?](https://anubislabs.ca/resources/what-is-governed-ai): Governed AI is AI designed with explicit controls for data access, evidence, review, permissions, policy enforcement, monitoring, and accountability.
- [Enterprise RAG Evaluation](https://anubislabs.ca/resources/enterprise-rag-evaluation): Enterprise RAG evaluation measures whether retrieval and generation are grounded, permission-aware, useful, current, and able to refuse weak evidence.
- [AI Audit Trails](https://anubislabs.ca/resources/ai-audit-trails): AI audit trails preserve the prompts, sources, tool calls, approvals, outputs, model versions, and reviewer decisions behind important AI-assisted work.
- [AI Policy Gates](https://anubislabs.ca/resources/ai-policy-gates): AI policy gates are deterministic checks and approval rules that decide whether an AI workflow can answer, act, escalate, or stop.
- [Human-in-the-Loop AI](https://anubislabs.ca/resources/human-in-the-loop-ai): Human-in-the-loop AI keeps people responsible for high-risk decisions while AI handles retrieval, drafting, comparison, triage, and evidence preparation.
- [AI for EPCM Workflows](https://anubislabs.ca/resources/ai-for-epcm): AI for EPCM workflows should focus on standards lookup, project evidence retrieval, proposal support, document review, and governed knowledge reuse.
- [AI for Proposal Teams](https://anubislabs.ca/resources/ai-for-proposal-teams): AI for proposal teams helps turn RFPs, resumes, project sheets, compliance matrices, and past pursuit material into source-backed response intelligence.
- [RAG vs Knowledge Graph](https://anubislabs.ca/resources/rag-vs-knowledge-graph): RAG retrieves source material for grounded answers, while knowledge graphs model entities, relationships, and structure that make retrieval more precise.
- [AI Document Review Controls](https://anubislabs.ca/resources/ai-document-review-controls): AI document review needs controls for source comparison, materiality, reviewer approval, version awareness, and traceable change summaries.
- [AI Readiness Checklist](https://anubislabs.ca/resources/ai-readiness-checklist): An AI readiness checklist helps teams decide whether a workflow has the data, ownership, controls, evaluation path, and deployment boundary required for a useful pilot.

## Case Studies

- [Enterprise Pursuit Intelligence Case Study](https://anubislabs.ca/case-studies/enterprise-pursuit-intelligence): Complex bid packages become structured requirements, risks, milestones, and source-backed proposal intelligence.
- [Evidence-Backed Staff Intelligence Case Study](https://anubislabs.ca/case-studies/evidence-backed-staff-intelligence): Resumes, project evidence, and curated profile data become proposal-ready expertise signals with source visibility.
- [Project Experience Retrieval Case Study](https://anubislabs.ca/case-studies/project-experience-retrieval): Structured portfolio filters and semantic search help business development teams retrieve prior project evidence.
- [Governed Corporate Knowledge Assistant Case Study](https://anubislabs.ca/case-studies/governed-corporate-knowledge-assistant): Internal standards, procedures, templates, and workflow knowledge become cited answers staff can trust.
- [AI-Assisted Document Review Case Study](https://anubislabs.ca/case-studies/ai-assisted-document-review): Document comparison workflows summarize meaningful changes while preserving human judgment and traceability.
- [Permit Intelligence Platform Case Study](https://anubislabs.ca/case-studies/permit-intelligence-platform): Municipal permit intelligence normalizes fragmented public data and extracts useful project signals.
- [Air-Gapped Sandbox Case Study](https://anubislabs.ca/case-studies/air-gapped-sandbox): Offline and VPC-ready AI runtime patterns for restricted workloads, strict egress, allowlisted tools, approvals, and auditable execution.
- [Industrial Visual Delivery Platform Case Study](https://anubislabs.ca/case-studies/industrial-visual-delivery-platform): CAD and BIM screenshots become decision-grade renders, videos, and delivery assets through a production AI workflow platform.
- [Early-Language iPad Companion Case Study](https://anubislabs.ca/case-studies/early-language-ipad-companion): A local-first iPad companion with AAC-style cards, parent-recorded voice, routines, phrase ladders, and child-safe speech practice.
- [Real-Life Mandarin Learning App Case Study](https://anubislabs.ca/case-studies/real-life-mandarin-learning-app): Offline-first spoken-language coaching with guided Mandarin practice, review gates, native audio, and real-world conversation flows.

## Machine-Readable Indexes

- https://anubislabs.ca/sitemap.xml
- https://anubislabs.ca/site-index.json
- https://anubislabs.ca/ai-index.md
- https://anubislabs.ca/.well-known/ai-site-index.json