Solutions Architect with 15+ years delivering enterprise systems across insurance, fintech, and SaaS platforms. Currently focused on production AI/agentic architectures — multi-agent orchestration, RAG pipelines, and human-in-the-loop systems on Azure. I design for observability, security, and governance from day one, with a strong preference for keyless, credential-free architectures that eliminate secret sprawl at the infrastructure level.
Reference implementations and tooling, each with a full documentation site.
| Project | What it is | Docs | Stars |
|---|---|---|---|
| mean-docker | Production-ready MEAN stack reference — MongoDB 8, Express 5, Angular 22 (zoneless), Node 24, fully containerized | Docs |  |
| clean-architecture-docker-dotnet-angular | .NET 10 Clean Architecture backend with an Angular 22 signals frontend — RBAC, activity logging, PostgreSQL | Docs |  |
| e-commerce-agents | Six specialized agents collaborating over A2A — dual-stack Python and .NET on Microsoft Agent Framework, with generative UI | Docs |  |
| ms-graph-mcp | Microsoft Graph MCP server — 85 tools across Outlook, Teams, OneDrive, SharePoint, Planner and Entra ID, delegated OAuth | Docs |  |
| mcp-generator | Agent readiness and governance layer for APIs — turn OpenAPI into a governed, portable MCP tool surface | Docs |  |
| ai-resources | AI Knowledge Hub — fundamentals through architecture patterns, agents and LLM systems | Docs |  |
- Multi-agent systems using Microsoft Agent Framework and LangGraph — orchestrating specialized agents for document processing, decision support, and process automation in enterprise contexts
- Agentic UIs with AG-UI and CopilotKit — building conversational interfaces that expose agent state and reasoning transparently to end users
- RAG pipelines on Azure AI Search + OpenAI, with retrieval evaluation, grounding verification, and content safety guardrails built in from the start
- Reference architectures for cloud-native .NET Aspire + Angular/Next.js applications following 12-factor and microservices principles
AI & Agentic Systems Microsoft AI Foundry · Azure OpenAI · Azure AI Search · LangGraph · Microsoft Agent Framework · RAG · Document Intelligence · Anthropic Claude · AI Governance & Content Safety
Cloud & Infrastructure Azure (AKS · Service Bus · Key Vault · App Service · Functions) · Docker · Kubernetes · Managed Identity · Workload Identity Federation · Private Endpoints · RBAC
Keyless Architecture Managed Identity over API keys · Workload Identity Federation · Entra ID service principals · credential-free service-to-service auth · zero-secret pipelines
Backend C# / .NET Core · .NET Aspire · Python · MediatR · Serilog · FluentValidation
Frontend Angular · Next.js · TypeScript · Tailwind CSS
Data Azure SQL · Cosmos DB · PostgreSQL · MongoDB
Observability OpenTelemetry · Azure Monitor · Application Insights · Distributed Tracing · Structured Logging
Security & Governance Zero Trust · OWASP · Azure Policy · Defender for Cloud · Compliance-aware architecture · Multi-tenant data isolation
DevOps Azure DevOps · GitHub Actions · Helm · Azure Container Registry
I use Claude Code for agentic development tasks — scaffolding, refactoring, and architecture exploration. GitHub Copilot for in-editor completion. All architectural decisions and production code go through human review. AI tools accelerate iteration; judgment still comes from experience.
I publish on architecture, AI integration patterns, and lessons from enterprise delivery at nitinksingh.com.