Skip to content

Technology

⚙️ BlueUP Platform: Layered Architecture

BlueUP is not a standalone product: it's an integrated platform in three layers designed so regulated organizations can adopt agentic AI without compromising security, compliance, or sovereignty.

Platform Architecture

┌─────────────────────────────────────────────────────────────┐
│            LAYER 3: GOVERNANCE & COMPLIANCE                 │
│                                                             │
│   ComplianceView         BlueUPALM (AML/DORA)              │
│   Continuous             Banking-grade regulatory           │
│   monitoring             compliance                         │
│   96 controls            Screening, AML, DORA               │
├─────────────────────────────────────────────────────────────┤
│            LAYER 2: SOVEREIGN EXECUTION                     │
│                                                             │
│   Rust financial engine  MCP / LMM Gateways                │
│   162k journals/sec      Tool and model                     │
│   Native multi-GAAP      governance                         │
├─────────────────────────────────────────────────────────────┤
│            LAYER 1: ZERO TRUST REACHABILITY                 │
│                                                             │
│   BlueUP Connect         OpenZiti / NetFoundry              │
│   Identity-first         Identity-first connectivity        │
│   desktop client         Dark services by default           │
├─────────────────────────────────────────────────────────────┤
│            SUBSTRATE: CRYPTOGRAPHIC IDENTITY                │
│                                                             │
│   Keycloak (humans) · SPIRE SVIDs (workloads)              │
│   Biscuit Tokens (offline authorization)                    │
│   OPA (infra policies) · Cerbos (app policies)             │
└─────────────────────────────────────────────────────────────┘

Layer 1: Zero Trust Reachability

The fundamental principle: without a valid cryptographic identity, no data path exists.

Services have no public IP, don't respond to port scans, don't appear on Shodan. They only "exist" for authenticated identities with matching policy.

ComponentFunction
OpenZitiOpen-source connectivity substrate: encrypted tunnels, dark services, service policy
BlueUP ConnectDesktop client showing users only their authorized services
Cryptographic identityEvery agent, service and human has a verifiable Ed25519 identity
Dark ServicesNo inbound ports, no public IP, invisible to the internet

Technology Partner: NetFoundry

Our connectivity substrate is built on OpenZiti, the open-source platform developed by NetFoundry. As an official NetFoundry partner, we offer both self-hosted deployment and managed connectivity for customers who require it.

NetFoundry is backed by investors including Cisco Investments and partners like Stellar Cyber and Intrusion for high-security environments.


Layer 2: Sovereign Execution

Business logic runs on controlled infrastructure with institutional-grade performance.

ComponentFunction
Financial engine (Rust)Multi-GAAP accounting (Sectoral, IFRS, Tax), 162k journals/sec
MCP GatewayGoverns which tools agents can invoke, by identity and policy
LMM GatewayControls access to language models with human approval points
gVisor sandboxKernel-level isolation per agent and service

Layer 3: Governance & Compliance

Continuous monitoring and regulatory compliance integrated by design, not bolted on.

ComponentFunction
BlueUPALMBanking-grade AML/DORA compliance: screening, incident management
ComplianceView96 controls aligned with NIST, ISO 27001, DORA, and FINOS. Automated collectors
OPACentralized access and infrastructure policy evaluation
Cerbos PDPContextual ABAC/RBAC authorization for business logic

Identity Substrate

The entire system shares a unified identity model:

LayerTechnologyFunction
HumansKeycloak OIDC/PKCEFederated authentication without static passwords
WorkloadsSPIRE SVIDs (Ed25519)Rotating cryptographic identity per service
AuthorizationBiscuit Tokens v6Capability tokens with asynchronous attenuation
EncryptionEnd-to-end mTLSMutual verification on every connection

Complete Technology Stack

LayerTechnologies
FrontendReact, TypeScript, CSS Modules
BackendRust (Axum), Python (FastAPI), NATS JetStream
SecurityOpenZiti, Keycloak, SPIRE, OPA, Biscuit Tokens, Cerbos
AI & DataVertex AI, MCP SDK, vLLM / Ollama (sovereign)
InfrastructureGoogle Cloud, Kubernetes (Talos Linux), Terraform, Gitea Actions
IsolationCilium (network), gVisor (kernel), eBPF (observability)

Design Principles

PrincipleDescription
Identity as PerimeterSecurity doesn't depend on server location, but on verifiable cryptographic identity
Dynamic PrivilegesAI proposes, but OPA policies and human intervention act as security brakes
Minimum Blast RadiusA compromise in one agent never translates to a systemic breach
Compliance by DesignDORA, AML, AI Act and GDPR are architectural requirements, not add-ons
Data SovereigntyProcessing never leaves controlled infrastructure

Want to dive deeper into the architecture?

Download our technical whitepaper on identity-first architecture for agentic AI.

Download whitepaper | Request demo

Last updated:

Zero Trust infrastructure for agentic AI in regulated industries