Enterprise AI · .NET Architecture · ERP

I connect your existing enterprise systems to production AI agent workflows — automating complex business processes that were previously impossible to scale.

15+ years of hands-on enterprise architecture in C#/.NET and ERP, combined with current production experience building AI agent systems: MCP integrations, multi-agent orchestration, Langdock. Deployed, not demo'd.

Let's talkSee approach

How I Work

🏗

Enterprise Architecture Depth

I understand business processes — supply chain, ERP, manufacturing, financial instruments — well enough to know what is actually worth automating. And what would create more problems than it solves.

⚡

Production AI Agent Engineering

MCP infrastructure, multi-agent orchestration on Azure AI Foundry, Langdock integration — built and running in production. Not slides. Not a whiteboard. Working systems that automate real workflows.

🎯

Advisory Model

I work selectively with a small number of clients. A few focused hours per week at a premium rate. The value is in architectural decisions and integration design — knowing what to build and how to connect it safely to what you already have.

🌐

Stakeholder Bridge

Ten years navigating complex international stakeholder environments — business users, technical teams, C-level, outsourced developers. I translate in both directions without losing meaning.

AI Agent Capabilities

Production experience — not prototypes:

Model Context Protocol (MCP):Built complete MCP infrastructure from scratch using JSON-RPC 2.0, enabling AI agents to dynamically query and act on enterprise system data. Integrated with Langdock and deployed to production.
Multi-agent orchestration:Designed and deployed multi-agent AI pipelines on Microsoft Azure AI Foundry — tool-call loops, automated delegation logic for complex business workflows. Replaces manual research and data enrichment processes end-to-end.
Enterprise system integration:Wired AI agents to external data APIs (company registries, market intelligence services) and internal databases for automated lead, client and candidate enrichment. Agents access live data, not stale snapshots.
Durable agent infrastructure:Database-backed agent state persistence replacing transient in-memory context — queryable, reliable research snapshots for long-running workflows. Enterprise-grade security with Entra ID / OAuth 2.0.
Microsoft ecosystem:Microsoft Copilot Studio integration for cross-platform agent capabilities. Real-time SSE and WebSocket streaming interfaces. Works within your existing Azure tenant.

Background

Experience

AI Integration Architect & Lead Solutions Architect2025–present · Remote

Designing and building enterprise AI agent systems: MCP infrastructure, multi-agent orchestration pipelines on Azure AI Foundry, agent-to-API integrations for automated data enrichment workflows.

Lead IT Consultant, Solutions Architect & Business AnalystJTI · 10+ years · Remote

De-facto lead architect and business analyst for a mission-critical custom ERP spanning JTI's globally integrated supply chain, manufacturing and logistics operations across multiple continents.

  • End-to-end ERP architecture: Japan integration (2022–24), Bangladesh (2020–21), factory operations, procurement, agro mobility
  • C#/.NET and MS SQL Server performance architecture: stored procedures, deadlock resolution, indexing — measurable throughput improvements
  • ETL pipelines, business logic automation, KPI calculation; financial instruments (umbrella loans, commodity-backed credit)
  • Mobile ERP extension: .NET Core backend + Vue.js PWA for handheld warehouse scanners via Azure API Gateway
IT ConsultantVarious clients

SAP & .NET ERP process specification (Spitalfields), tournament scheduling platform (myLudus), earlier C#/.NET development and team leadership.

Expertise

AI Agent EngineeringModel Context ProtocolC# / .NETERP ArchitectureMS SQL ServerMicrosoft AzureETL & Data PipelinesSupply ChainStakeholder ManagementBusiness Analysis

Education

Diplom-Engineer (Dipl.-Ing.)
Electrical Engineering, Computer Science
University of Banja Luka · 2009

Languages

English (C1/C2) · German (C1) · BCS (native)

Location

Fürth, Germany 🇩🇪

Full CV available upon request. Get in touch →

Let's Talk

Working on connecting AI agents to your existing systems? Need an architecture review or integration design? I'd like to hear about it.

📅Book a meeting💼LinkedIn✉️Email me