01
PLATFORM LAYER
Application Layer
Your enterprise applications, Spring Boot microservices, APIs, and business workflows consuming intelligence.
Spring Boot 3/4MicroservicesREST EndpointsDomain Logic
↗
←Back to Shree AI OS
SHREE AI OS · PLATFORM
01An operating
An operating
foundation for
intelligent software.
Shree AI OS is a platform for building intelligent software with reusable AI infrastructure, explicit architectural boundaries, stable interfaces, and a runtime designed to connect intelligence with execution.
v1.0.6 · DEVELOPER PREVIEW
01 · THE PLATFORM
Not an application.
A foundation.
Modern intelligent applications repeatedly rebuild many of the same foundations — memory, context, reasoning, planning, execution, runtime services, and integration infrastructure.
Shree AI OS approaches this problem from the platform layer. Instead of making every application rebuild intelligence infrastructure independently, the platform provides reusable foundations that applications can build upon.
THE PLATFORM PRINCIPLESHREE / 01
S
Build intelligent systems
on reusable foundations.
The platform separates infrastructure from applications, allowing intelligence capabilities to evolve independently while applications remain focused on their own domains.
02 · PLATFORM ARCHITECTURELayers designed
Layers designed
to work together.
Shree AI OS is structured as a layered platform. Each layer has a defined responsibility while remaining connected through explicit interfaces and architectural boundaries.
02
PLATFORM LAYER
SDK Layer (10 Facades)
Public Java 21 client interfaces providing typed, deterministic access to memory, planning, code patches, and model routing.
MemorySDKKnowledgeSDKPlanningSDKReasoningSDKReflectionSDKIdentitySDKExecutionSDKProjectSDKDeveloperSDKMultiAgentSDK
↗
03
PLATFORM LAYER
Runtime Orchestration
The deterministic execution brain enforcing the 11-stage pipeline, fail-closed authorization, multi-tenant boundaries, and event dispatching.
11-Stage PipelineFail-Closed GateIntent RouterEvent BusLlmRouter
↗
04
PLATFORM LAYER
Kernel Services
In-process cognitive kernels for episodic memory, pgvector hybrid RRF search, topological DAG planning, and AST code patch generation.
Episodic Memorypgvector HNSW + GINK0.6 AcquisitionDAG PlannerAST Engine
↗
05
PLATFORM LAYER
LLM Provider Layer (BYOK)
Swappable model integration with automatic HTTP 429/503 exponential backoff retries and deterministic in-memory fallback.
Google GeminiOpenAIOllamaIn-Memory Fallback
↗
03 · THE FLOW
From application
to intelligence.
Applications interact with Shree AI OS through developer interfaces. The platform then connects those requests with runtime services and reusable intelligence capabilities.
01Your ApplicationJava 21 / Spring Boot 3/4
↓
0210-SDK Facadesclient.chat() / memory() / planning() / developer()
↓
0311-Stage PipelineIdentity → Context → Memory → Knowledge → Reasoning → Inference → Planning → Execution → Reflection → Store → Review
↓
04Dual-Mode SynthesisStrict RAG Citations vs. General Fallback
↓
05K0.6 Knowledge & BYOKDomain Isolation & Multi-Provider Router
04 · WHY A PLATFORM
Stop rebuilding
the same intelligence.
A platform approach creates a shared foundation that can serve multiple intelligent applications while keeping application concerns separate from infrastructure concerns.
Reusable
Intelligence capabilities can be reused across multiple applications instead of being rebuilt independently.
Modular
Platform capabilities remain separated into clear layers and components with explicit responsibilities.
Extensible
The architecture is designed so future capabilities can be introduced without redefining the entire platform.
Developer-focused
Developers interact with stable interfaces rather than depending directly on internal implementations.
PLATFORM → APPLICATIONThe platform
The platform
comes first.
Applications are where the platform becomes useful. They demonstrate what can be built when intelligence infrastructure is available as a reusable foundation.
Explore Applications↗