Systems that
Designed and built.
Product architecture, intelligent workflows, semantic systems and scalable infrastructure — from early concept to production. For founders, innovation teams and operators who need execution, not slide decks.
Four practices, one integrated approach.
Not services offered in isolation — systems built across all four layers simultaneously, because that's how operational problems actually work.
Operational AI Systems
AI systems that reduce coordination overhead, automate workflows and integrate into real operational environments — not isolated demos.
AI-native Product Infrastructure
Scalable platforms, semantic systems and monetization architectures built for long-term operational growth — from MVP to production.
Adaptive Product Systems
Conversational systems, AI-assisted UX and interaction models that connect humans with intelligent workflows and operational data.
Creative Systems & Media Infrastructure
Voice, sound and generative media systems designed as scalable production infrastructure — not isolated creative assets.
How systems get built
Each engagement starts with a real problem. These are the system types I architect and build — without client names, focused on the thinking.
Conversational AI Project Creation System
Complex project intake relied on fragmented communication, repeated qualification and disconnected handoffs.
Conversational interfaces that guide users through project creation:
- —AI-assisted requirement gathering
- —Contextual memory handling
- —Intelligent question flows
- —Automated project structuring
- —Dynamic workflow routing
- —Recommendation systems
- AI-native project workflows
- Replaced fragmented communication
- Faster intake cycles
Semantic Search & Recommendation Systems
Large databases returned imprecise results relying solely on traditional keyword filtering.
AI-powered search systems using:
- —Embeddings-based similarity search
- —Semantic ranking
- —Contextual filtering
- —Multilingual indexing
- —Recommendation logic
- —Intelligent discovery systems
- Intent-driven retrieval
- Higher result relevance
- Multilingual coverage
AI Workflow & Operations Systems
Repetitive coordination tasks consumed team capacity that should have been focused on delivery.
AI-assisted workflow infrastructure for:
- —Task orchestration
- —Project coordination
- —Communication routing
- —Contextual knowledge aggregation
- —Production management
- —Operational automation
- Reduced coordination overhead
- Intelligent task routing
- Scalable operations
AI Agent & Automation Systems
Complex operational workflows required repeated human decision-making that could be delegated to intelligent systems.
AI-agent workflows capable of:
- —Delegated task execution
- —Contextual memory routing
- —Workflow orchestration
- —Local and cloud AI integration
- —Multi-agent coordination
- —Automated operational support
- Delegated complex tasks
- Multi-agent coordination
- Operational AI support
AI-assisted Talent & Freelancer Matching
Manual talent selection was slow, subjective and failed to surface the best matches from large candidate pools.
Intelligent matching systems through:
- —Semantic profile analysis
- —Contextual recommendations
- —AI-assisted filtering
- —Conversational project intake
- —Ranking systems
- —Workflow-integrated matching logic
- Higher match quality
- Reduced selection time
- Intent-driven discovery
How I think about systems.
These aren't methodologies. They're constraints I've arrived at from building things that failed when I ignored them.
Systems over isolated features
Most business bottlenecks are coordination problems in disguise. The fix is rarely a new feature — it's a better-connected system.
AI must reduce friction, not add it
AI only creates value when it integrates into real operational workflows. A model that runs in isolation doesn't count.
Interfaces are operational tools
UX is not decoration. How a system is experienced determines decision speed, trust and actual adoption.
Human-in-the-loop where trust matters
Full automation is not always the right goal. The right architecture keeps humans in control where accountability is non-negotiable.
Automation should reduce cognitive load
Good automation removes complexity from the people using it — not just moves it to a different layer.
Scalable systems need reusable primitives
Infrastructure built to enable future products is worth more than infrastructure that just solves today's problem.
Work
Featured projects and full archive — filtered by discipline.
Voicfy Platform
Built a full-stack voice platform from product concept to production deployment, including billing, user flows and scalable architecture.
Created a reusable platform foundation for rapid feature rollout and monetization.
Voicfy Real Voice TTS (Diffusion)
Developed a diffusion-based real voice TTS system with training and inference pipelines for production usage.
Improved output quality and enabled reliable high-volume voice generation workflows.
Movie Score for Treppe Aufwärts
Composed and produced the movie score for Treppe Aufwärts, a German feature drama (2015) by Mia Maariel Meyer. The music direction was built around scene tension, family conflict and editorial rhythm, from thematic sketching to final mix delivery.
Established a coherent sonic identity across key character arcs and supported emotional continuity between dialogue-driven and high-pressure sequences. The score approach helped connect narrative pacing from festival presentation to theatrical release in Germany.
KaDeWe Group Digital Design
Designed digital experience assets and visual systems for KaDeWe Group brand and campaign touchpoints.
Improved visual consistency across premium retail communication formats.
AI Operations Copilot
Implemented triage and routing automation across service channels with human-in-the-loop controls and audit-ready execution logs.
Reduced manual handling effort and stabilized operational throughput.
National Geographic / Nat Geo Wild
Produced and coordinated multilingual announcer and voice recording sessions for Nat Geo Wild and National Geographic international programming from 2008 to 2010, covering studio production, language versioning, delivery support and on-air presentation workflows for wildlife documentaries and factual entertainment formats.
Consistent multilingual broadcast delivery across voice talent coordination, localization workflows and final transmission standards.
From concept to production — systems that run.
The work was never shaped by a single discipline.
Entrepreneurship, media production, motion design, sound, technology, electronics and software systems developed in parallel from an early stage — eventually converging into a systems-oriented way of thinking.
That interdisciplinary foundation shaped an approach where products, workflows, interfaces, infrastructure and human experience are treated as connected parts of the same system.
Today the focus lies on building AI-native operational systems, intelligent workflows and scalable digital products that bridge technology, automation, UX and creative direction. The strength is not specialisation in one discipline — it's connecting disciplines that are usually separated.
Tools & technologies
Building AI systems that run in real processes — not in demos.
The focus is on operational impact: workflows that actually get automated, systems that integrate into the team, and infrastructure that keeps running after the engagement ends.
What this covers
- Agentic workflows: routing, classification, escalation and task execution across internal tools.
- AI-native product infrastructure: LLM integration, retrieval systems, evals and observability.
- Automation of manual processes with human-in-the-loop where accountability matters.
- From assessment to running system — implementation, integration and handoff to your team.
Process
Assessment
We review your workflows, identify manual steps and find the processes with the biggest AI potential.
Pilot
Once a use case is clear, an initial version is built with real data and a clean feedback loop.
Rollout
Then comes integration into your systems, production monitoring and handoff to your team.
Let's discuss your architecture.
For teams building operational AI systems, scalable products and intelligent workflows.
- AI workflow transformation
- Conversational product systems
- Semantic search infrastructure
- AI-native SaaS platforms
- Operational automation
- Voice AI systems
- Fractional product & AI leadership