Skip to main content
Backbase hosts and operates the Agentic Platform on Azure, providing production-ready infrastructure for building and deploying AI agents. This architecture uses Kubernetes, GitOps, and enterprise-grade observability, with a 99.5% or 99.9% availability service level agreement (SLA).

​
Architecture overview

Agentic Platform architecture diagram

​
Platform layers

The architecture diagram shows five foundational layers that work together:
  • Kubernetes: Container orchestration on Azure Kubernetes Service
  • Service Mesh: Istio for traffic management, security, and observability
  • Container Runtime: Docker for containerization
  • Packaging: Helm charts for deployment configuration
  • Vector Storage: PostgreSQL with pgvector for embeddings and RAG
  • Caching: Redis for session state and fast lookups
  • Object Storage: Cloud-native storage for artifacts and models
  • Message Bus: Sync Hub provides Azure Service Bus for async workflows
  • Registries: Azure Container Registry (ACR) and GitHub Packages
  • Provider Abstraction: Unified interface to Azure AI Foundry, OpenAI, Gemini, Anthropic, or BYO models
  • Policy Enforcement: Guardrails, rate limiting, cost management, and content safety
  • Traffic Management: Semantic caching, load balancing, and intelligent request routing
  • PII Protection: Automatic detection and sanitization
  • SDK Access: BB AI SDK for standardized integration
  • Agent Workloads: FastAPI-based APIs with background workers
  • Orchestration: Agno and LangGraph for multi-step workflows
  • MCP Integration: Model Context Protocol servers for tool access
  • Banking Services: Pre-integrated domain services (deposits, payments, loans, fraud)
  • Guardrails and Safety: Real-time evaluations, content safety filters, PII detection/sanitization, jailbreak prevention, prompt validation
  • Observability: OpenTelemetry traces, Langfuse runs, metrics, and logs
  • Ingress: Azure API Management (APIM) with DNS, WAF, SSL termination
  • GitOps: Argo CD for declarative, automated deployments
  • CI/CD: Automated pipelines with PR checks, builds, and releases
  • Self-Service: Repository provisioning and infrastructure automation
  • Monitoring: Grafana dashboards, PagerDuty alerts, real-time evaluations

​
Runtime flow

Agentic Platform runtime request flow
When a client makes a request:
  1. Client → APIM → Agent API: Request enters through APIM with authentication and rate limiting
  2. Agent → AI Gateway: Agent calls LLM through gateway with guardrails and traffic control applied
  3. AI Gateway → LLM Provider: Request routed to selected provider (Azure AI Foundry or BYO) with policies enforced
  4. Tool Execution: Agent calls MCP servers (banking domains) or direct REST/GraphQL APIs
  5. Observability: OTel traces and Langfuse runs capture every step; logs and metrics flow to platform sinks
  6. Real-time Evaluation: Automated evaluations flag risks (safety, PII, jailbreaks) and feed continuous improvements

​
Next steps

Technology stack

Explore the detailed technology stack and implementation deep dive

Data protection and backup

Automated PostgreSQL backups, retention, and restore requests

Get started

Start building your first AI agent on the platform