AI Decision Orchestration Service

A production-grade microservices orchestrator built with Go and React, focusing on Observability, Resilience, and Traceability.

🚀 Features

• Centralized Logging (MDC-style): Context-aware logging using slog (automatically enriches logs with correlation_id and user_id).
• Distributed Tracing: OpenTelemetry integration with Jaeger support.
• Metrics: Prometheus metrics endpoint (/metrics) with custom vectors.
• Resilience: Circuit Breakers, Retries, and Timeouts for all downstream calls.
• Docker Ready: docker-compose setup for the entire stack.
• Kubernetes Native: Helm charts for deployment on any cluster.

🛠️ Getting Started

Prerequisites

• Go 1.23+
• Docker & Docker Compose
• Kind (for local Kubernetes)

Running with Docker Compose

docker-compose up --build

The API will be available at http://localhost:8080.

Running Locally

1. Start the simulator: go run cmd/simulator/main.go
2. Start the server: go run cmd/server/main.go

📊 Observability

• Prometheus: http://localhost:9090
• Jaeger UI: http://localhost:16686
• Metrics: http://localhost:8080/metrics

🏗️ Architecture

The service follows a clean architecture pattern:

• internal/api: HTTP Handlers
• internal/orchestration: Core Decision Logic
• internal/clients: Downstream microservice clients
• internal/middleware: Resilience and Observability layers

☸️ Kubernetes

Deploy to a local Kind cluster:

kind create cluster --config kind-config.yaml
helm install orchestrator ./charts/orchestrator

📝 License

MIT