Building intelligent platforms that solve real-world problems in healthcare, enterprise automation, and human-machine collaboration.
📍 Germany | 🕐 CET/CEST
I'm a systems architect and AI strategist passionate about creating intelligent systems that augment human capabilities in healthcare and enterprise environments. My work focuses on:
- Medical AI: Democratizing access to medical expertise through AI-powered diagnostics
- Multi-Agent Systems: Building collaborative AI teams that solve complex problems
- AI Agent Technologies: Developing Agent Skills, Tools, Plugins, Agent Harnessing
- Ethical AI: Ensuring responsible, secure, and privacy-preserving AI implementations
"We don't just develop software. We build systems that think."
A modular skill library that gives AI agents clinical‑grade radiology workflows for imaging analysis, PACS/DICOM integration, reporting, patient communication, and research.
🔗 agent skills • [Live Demo - coming soon]
Key Features:
- Modular skills architecture that agents can load individually or as full workflows including plugins and tools.
- End‑to‑end radiology pipeline coverage: imaging, reporting, communication, quality, research.
- PACS/DICOM/HL7‑FHIR integration patterns for real clinical environments.
- AI‑augmented workflows including detection tools, report QC, and dataset prep.
- Corpus Analytica–verified clinical standards ensuring safety, consistency, and interoperability.
HALO Core is an intelligence workspace for teams that need to turn raw information into clear, usable outputs.
Key Features:
- Team-agent capable - run specialized AI roles with delegation strategies.
- Complexity-behind-glass - transparent execution, inspect tool calls, agent actions, and traces when needed.
- Source-grounded - answers are driven by selected evidence.
- Modular Design - Specialized agents with Agno and OpenAI Responses API
- Scalable Infrastructure - Auth0, JAMstack, vector databases (LanceDB, PGvector)
- Knowledge Systems and AI Memory - RAG solutions, knowledge graphs, semantic search, contextual understanding
- Business Automation and Workflow-first - Process optimization, built for end-to-end work, not one-off prompts, e.g. generate reports, infographics, and data tables.
A Streamlit‑based, agent‑driven medical AI app that performs intelligent diagnostics, medical image analysis, and research retrieval through a secure, intuitive interface.
Key Features:
- Medical Imaging Agent for structured analysis of X‑ray, CT, MRI, ultrasound images, all images including mobile phone camera shots.
- Modern Streamlit UI with light/dark themes, real‑time streaming, and configuration pages.
| Medical AI | Agent Systems | Backend Architecture | Frontend Development |
|---|---|---|---|
| DICOM/PACS Integration | Multi-Agent Orchestration | Microservices Design | Streamlit Applications |
| FHIR/HL7 Standards | RAG Implementation | Vector Databases | Responsive Web Design |
| Medical Imaging Analysis | Knowledge Graphs | API Design | User Experience |
| Healthcare Compliance | Workflow Automation | Database Architecture | Interface Design |
"We don't just develop software. We build systems that think."
Bernhard Zechmann
Founder of CorpusAnalytica & watunga.ai
