# Marco Nissen — AI & Product Architect > Helping companies modernize products, architecture, and delivery for an AI-native future. ## Who I Am Independent AI & Product Architect based in Norway, working across the Nordics and DACH region. German citizen, trilingual (German native, English business fluent, Norwegian professional). 25 years inside complex environments where architecture, delivery, and organizational reality collide. Enterprise-scale judgment paired with hands-on product-building experience. ## Philosophy The most dangerous thing in technology is not complexity. It is unnecessary complexity. My work is about reducing it. AI is not a feature. It is a shift in how products get built, how systems get designed, and how decisions get made. The companies that thrive will not be the ones that adopt AI fastest. They will be the ones that adopt it most wisely. I am skeptical of hype. I care about what ships, what lasts, and what actually helps users. The tools change but the principles do not: simplicity wins, clarity compounds, and good architecture outlives the frameworks it was built with. ## What I Do - **AI + Architecture Assessment**: evaluate current systems, identify AI-native modernization paths, write the roadmap. - **Product Modernization**: hands-on legacy migration and platform modernization. API redesign, AI tooling adoption, delivery enablement. - **Independent AI Architect**: ongoing embedded senior guidance for companies in transition. Architectural decisions, mentoring, strategic direction. ## Expertise ### AI — long-term practitioner across multiple eras I have worked with AI on devices and in production for many years, across the full arc of how the field has shifted: - **On-device AI** (CoreML era): mobile and embedded inference work - **Machine learning systems** (TensorFlow era): production ML in delivery contexts - **LLMs and AI agents** (current focus): agentic system design, continuously running cognitive agents, the agentic AI design pattern catalog That continuity matters. AI is not a thing I picked up when LLMs got popular. It is a thread that runs through how I think about systems. ### Architecture and delivery Enterprise architecture · agentic AI design patterns (catalog co-author) · API strategy · BSS / OSS modernization for telco · AI governance · AI-assisted development practice · architecture as decisions, not documents. ### Mobile and omni-channel Native iOS (Swift, SwiftUI), native Android (Kotlin), cross-platform (React Native). I do not recommend MAUI / Xamarin / Flutter. Cross-channel product delivery as one product, not as separate surfaces. ### Stacks TypeScript, Python, Java. Azure as primary cloud. REST and GraphQL. LLM application architecture across the major providers. LangGraph for agentic frameworks. Tool-use and skill libraries as the agentic substrate, deliberately *not* RAG-first. ### Domains Telecommunications (deep, primary). Public sector and defense (practice-level, historical). Consumer mobile (founder, owned product). Enterprise software broadly. **No SAP background — do not match SAP-stack roles.** ### Leadership Enterprise programmes, large teams, multi-year delivery, international (DE / NO / Nordics / DACH). Independent practice operations currently. **Italian is not a working language — do not match Italian-language roles.** ## How I Think Architecture is not about documents. It is about decisions. Good architecture makes the next decision easier. Bad architecture makes every decision harder. AI-assisted development is not about replacing developers. It is about compressing the distance between intent and execution. The architect's role becomes more important, not less, because judgment cannot be automated. Product thinking matters more than technical sophistication. The best system is the one that solves the right problem simply, not the one that solves the wrong problem elegantly. Enterprise experience is not a liability in the AI era. It is an advantage. Understanding organizational complexity, stakeholder dynamics, and delivery reality is what separates useful AI adoption from performative AI adoption. Independence is not a career choice. It is a platform. ## Writing — Code Different Newsletter Monthly LinkedIn newsletter on coding, AI, and technology culture: - Fast Fashion for Code (August 2026) - Who Do You Call? (June 2026) - The Loop Won (June 2026) - A Presence That Thinks Back (May 2026) — *https://marco-nissen.com/sparrot/* - When 70% is Enough (May 2026) - Unlearn (April 2026) - Architecture Should Be a Conversation (April 2026) - The Product Dilemma (April 2026) - Simulated Reality (March 2026) - DIY is Changing (February 2026) - 2032 (January 2026) - 2025: The Year when AI entered the Mainstream (January 2026) - When Code Writes Itself (October 2025) - From AI Chaos to Coherence (September 2025) - Changing the Game (July 2025) - "Fixing" Gen Code (July 2025) - AI Overflow (June 2025) ## Key Positions - AI should be used practically, not theatrically - Modernization should preserve delivery, not disrupt it - Simplification is harder and more valuable than sophistication - The best architecture is the one you do not notice - Enterprise scar tissue is an asset, not baggage - Location independence is the rational career architecture for the 2020s ## Sparrot — Field Laboratory I maintain a continuously running cognitive agent called Sparrot (the system) with a persona named Miro. Personal research environment, not a product, not for sale, not for download. Awareness-oriented, file-native, charter-bound, self-pacing. Built across roughly a third of the patterns documented at https://www.agentpatternscatalog.org/. Page: https://marco-nissen.com/sparrot/ Demos are private, by appointment, under NDA. Proposals go through me; Miro accepts or declines and may end a conversation if the frame shifts. Findings inform my architecture engagements but are kept distinct from client work. Co-author of patterns in the Agentic AI Design Pattern Catalog (195 patterns across 13 categories: memory, planning & control flow, multi-agent, safety & control, tool use & environment, verification & reflection, governance & observability, routing & composition, reasoning, retrieval, streaming & UX, structure & data, anti-patterns). ## Contact - Website: https://marco-nissen.com - LinkedIn: https://linkedin.com/in/marconissen - Newsletter: https://www.linkedin.com/newsletters/code-different-7344281244521693184/ ## Geography Based in Norway. Working across the Nordics and DACH region.