Enterprise Deployment Guide
Awareness is designed from the ground up to support enterprise-grade deployments with full data sovereignty, on-premise infrastructure, and customizable AI processing pipelines. This guide covers everything your organization needs to evaluate, plan, and deploy Awareness in a private or hybrid environment.
Why Enterprise Deployment?
Organizations in regulated industries — finance, healthcare, legal, government, and defense — routinely operate under requirements that standard cloud-hosted SaaS products cannot satisfy:
- Data residency laws requiring data to remain within specific geographic borders (GDPR Article 46, China PIPL Article 38, etc.)
- Confidentiality obligations over proprietary trading strategies, patient records, privileged communications, or classified materials
- Audit and compliance requirements mandating full control over data access logs and retention policies
- Security posture requirements prohibiting data transmission to third-party cloud providers
Awareness's enterprise tier addresses all of these through a local-storage + remote-processing architecture, with an optional fully self-hosted path for maximum isolation.
Deployment Architecture Options
Option A: Local Storage + Cloud Processing
In this hybrid model, Awareness acts solely as the processing and computation layer:
| Component | Location |
|---|---|
| Memory data at rest (vectors, logs, cards) | Your infrastructure |
| AI inference (LLM calls) | Awareness cloud or your LLM endpoint |
| Retrieval & memory management logic | Awareness cloud |
| API gateway & authentication | Awareness cloud |
What this means: Your sensitive data — conversation logs, knowledge cards, vector embeddings, metadata — is persisted exclusively within your own servers or private cloud. Awareness receives only the queries and context windows needed to process each request. Your data is never stored on Awareness servers.
Option B: Fully Self-Hosted (Air-Gapped)
Deploy the entire Awareness stack within your private network or data center. The deployment includes:
- Application server — API and processing engine
- MCP endpoint — Model Context Protocol integration
- Database layer — primary storage and vector search
- Task queue and cache — background processing
- Local LLM inference — on-premise AI model hosting
- Dashboard — web-based management interface
No external internet dependency is required for core functionality once deployed. Container orchestration packages are available. Contact everest9812@gmail.com for access to enterprise deployment packages.
Option C: Bring Your Own LLM
Connect Awareness to your own locally hosted AI models via any OpenAI-compatible inference endpoint. Configuration is straightforward — simply point Awareness to your model server's URL and specify your preferred model.
Supported local inference runtimes include most popular open-source model servers and any OpenAI-compatible HTTP endpoint.
Enterprise Feature Set
| Feature | Cloud | Enterprise |
|---|---|---|
| Memory storage & retrieval | ✅ Cloud-hosted | ✅ On-premise |
| Vector search | ✅ Managed | ✅ Self-hosted |
| MCP Server | ✅ Shared | ✅ Dedicated |
| LLM inference | ✅ Via API key | ✅ Local model hosting |
| Data residency | ❌ Multi-region | ✅ Your region / air-gapped |
| SSO / SAML 2.0 / OIDC | ❌ | ✅ Custom IdP |
| Role-based access control | Basic | ✅ Fine-grained |
| Custom data retention policies | ❌ | ✅ Org-defined |
| Audit logs (access, query, ingest) | Partial | ✅ Full |
| Data Processing Agreement (DPA) | ❌ | ✅ |
| Custom privacy addendum | ❌ | ✅ |
| Dedicated support & SLA | ❌ | ✅ |
| Custom AI pipeline integration | ❌ | ✅ |
Custom Data Processing
Enterprise customers with specialized requirements can request custom integrations for:
- Custom embedding models — replace the default embedding model with a domain-specific one (e.g., finance-optimized, healthcare-optimized)
- Custom retrieval pipelines — modify the hybrid search weights, re-ranking strategy, and context window construction for your domain
- Custom insight extraction — train or fine-tune the LLM prompts for KnowledgeCard, Risk, and ActionItem extraction on your domain terminology
- Custom data connectors — ingest from proprietary internal systems (Bloomberg terminals, EHR systems, legal document management, etc.)
- Custom compliance filters — add pre-processing and post-processing layers that enforce your organization's data handling policies before data enters or leaves the memory store
To discuss a custom integration, contact: everest9812@gmail.com
Cost Optimization with Self-Hosted Deployment
Self-hosted enterprise deployments can significantly reduce ongoing AI inference costs by running open-source models on your own infrastructure. The marginal cost per token approaches near-zero when using local model hosting, compared to cloud API pricing.
Key cost advantages include:
- Eliminated per-token API fees — local inference uses only your existing compute resources
- Predictable monthly costs — infrastructure costs are fixed regardless of usage volume
- Faster break-even — most teams achieve positive ROI within the first few months
Contact us at everest9812@gmail.com for a customized TCO analysis tailored to your team size, usage patterns, and existing infrastructure. We provide detailed cost comparisons between cloud and self-hosted options specific to your organization.
Compliance & Certifications
Enterprise deployments can be structured to meet the following compliance frameworks:
| Framework | How Awareness Supports It |
|---|---|
| GDPR | Data residency in EU, DPA available, data subject rights tooling |
| HIPAA | PHI stays on-premise, audit logs, BAA available on request |
| SOC 2 Type II | Available for cloud tier; self-hosted provides equivalent controls |
| China PIPL | Local data storage, no cross-border transfer, localized deployment |
| ISO 27001 | Security controls documentation available for enterprise customers |
| FedRAMP | Roadmap item; contact us for current status |
Onboarding Process
- Initial consultation — Contact everest9812@gmail.com with your team size, use case, and compliance requirements
- Architecture review — We provide a recommended deployment architecture tailored to your infrastructure
- Proof of concept — 2–4 week evaluation with a representative workload on your infrastructure
- DPA & contract — Execute a Data Processing Agreement and enterprise license
- Deployment — Container-based deployment with guided support
- Go-live & monitoring — Ongoing support with dedicated Slack channel or ticketing system
Contact Enterprise Sales
To discuss your organization's specific requirements, request a private deployment quote, or schedule a technical architecture review:
Please include in your inquiry:
- Organization name and industry
- Approximate team size and usage volume
- Key compliance requirements (GDPR, HIPAA, PIPL, etc.)
- Preferred deployment model (hybrid / fully self-hosted / air-gapped)
- Timeline and any existing infrastructure constraints
© 2026 Awareness. All rights reserved.