Markdown Experts in Germany

• Development of an agent-based governance platform for the automated assurance of EU AI Act compliance and ODA-compliant orchestration of AI services in complex enterprise environments.

• Design and implementation of an agent-based "Mission Control" framework (Aletheia Conductor) for autonomous state monitoring and process control.

• Development of "Compliance-as-Code" (CaC) solutions based on OPA/Rego for system-wide enforcement of regulatory guardrails.

• Integration of TM Forum ODA standards (TMF630, TMF622, TMF642) to ensure interoperability and standardization.

• Building a highly available event-driven architecture using Redpanda and CloudEvents v1.0 for near-real-time event processing.

• Implementation of an audit-proof "Evidence Chain" through cryptographic linking of trace logs in preparation for automated audits.

• Tech Stack: Java 21 (Quarkus Native), TypeScript (Next.js), Redpanda (Kafka API), CloudEvents v1.0, OPA (Open Policy Agent) & Rego, TimescaleDB, ZincSearch, Redis, TM Forum ODA, Git, GitHub, Clean Code Development, Like-C4.

Multi-Agent Document-Generation Platform

Problem:

Turn a single prompt into a well-researched, structured and cited document, while keeping the user in control: every stage can be reviewed, corrected or revisited before the system moves on.

Architecture:

A multi-agent system orchestrated by a central controller. More than 20 specialised pydantic-ai subagents (understanding, structuring, planning, research, writing, assembly, cross-reference check, glossary, citation, reference table and more) pass results forward through a shared plan persisted in Postgres (asyncpg); a runner layer dispatches each task by name to its agent, and mechanical steps such as formatting, table of contents and list of figures are deterministic code instead of LLM calls. Provider (Anthropic or OpenAI) is a configuration setting, and responses stream over Server-Sent Events.

Hosting & Infrastructure:

Stateless container run as a Scaleway Serverless Container on a private network with managed PostgreSQL, object storage for finished documents and assets, and centralized log shipping (Loki / Scaleway Cockpit), provisioned with OpenTofu modules and deployed through GitLab CI; usage is metered per API key.

My Contribution & Key Decisions:

• Central decision controller: a satisfaction classifier turns the user's free-text feedback into a verdict (keep working, advance, cancel or revise an earlier task), and a middleware applies it to the plan. Plan progression supports jumping back to earlier tasks and re-running the stages after them.
• Research and writing fan-out: planning produces a plan from which a research-design agent generates individual research tasks and a writing-design agent generates individual writing tasks, each linked to the research it needs. Every task runs in its own dedicated agent with its own conversation history.
• Self-review and human-in-the-loop: after each step a critical-thinking agent reviews the output; the user can accept, refine, or ask questions about a critique before the plan advances.
• MCP integrations: two remote MCP servers, web search and internal document search, connected through pydantic-ai with retries and long-running tool-call timeouts. Classifier agents decide per task whether web search is needed, and a small model condenses raw search results into structured findings before they reach the writing agents.

Usage Metering:

Clients authenticate with an API key validated against a key-management service; consumption is charged in credits calculated from token usage per model plus a markup, and exhausted quota is answered with HTTP 429.

• Development of an AI-powered project organization tool for a civil engineering company that intelligently links project, task, tender, schedule, and document data through a knowledge graph.
• Implementation of AI features for document analysis, information extraction, context-based assistance, and voice-based data capture based on Microsoft Azure AI, reducing administrative effort, making information available faster, and supporting project teams in decision-making.
• Tech stack: Python, React, TypeScript, FastAPI, Claude Code, Codex, Graphify, PostgreSQL, Microsoft Azure AI Foundry, Azure OpenAI, Azure AI Speech, Azure AI Document Intelligence, Microsoft Graph, Microsoft Entra ID, Docker, Git, CI/CD.

• Recurring problem in my own AI-assisted projects: requirements analysis, use cases, and architecture decisions can be created quickly with AI support, but remain difficult to follow and scattered across Markdown files – knowledge is lost as soon as it is no longer in the context window. arknet turns requirements engineering and architecture knowledge into structured, verifiable data instead of plain text: requirements, use cases, and architecture decisions form a consistently linked knowledge graph, traceable from requirement to architecture decision – queryable by both people and AI agents. Technically based on RDF/OWL and a custom MCP server.

• Result: Working MCP daemon, Docker image published automatically to GHCR, nine hexagonal modules, eleven ADRs (including an Open-Core licensing model). Requirements engineering and Ubiquitous Language hexagons are active. Public as a Community Edition under Apache-2.0 since 07/2026 (github.com/kogn-io/arknet), together with the Claude Code plugin and GHCR image; Open-Core model.

• Label: Java, Maven, RDF, RDF4J, OWL, SPARQL, Model Context Protocol, Spring AI, Docker, GitHub, Git, Claude Code, Obsidian, DDD, Hexagonal Architecture, ArchUnit, JUnit, AssertJ, Interface Development, Software Architecture, Continuous Integration, Knowledge Management

• Advising companies on building, migrating and optimising BI and reporting landscapes (Power BI, SQL, Python, ETL)

• 5 client engagements in real estate and finance since 05/2025: taking over and stabilising existing reporting, automating recurring standard and management reports, building cash-flow models

• Proposal and feasibility assessments for BI and reporting projects

• Using AI-assisted development (Claude Code) to accelerate automation, tooling and web/app development

• Custom ERP system

• Problem: A client's core processes ran on scattered, siloed Excel files with no central data storage – error-prone, hard to scale and impossible to analyse end-to-end.

• Approach: Captured the business processes and requirements, modelled the data and developed iteratively together with the business team.

• Implementation: Built a tailored, web-based ERP system with a central database, role-based modules and automated reporting – delivered using AI-assisted development in Claude Code.

• Timesheet app

• Starting point: Time tracking based on an overgrown, macro-heavy Excel template – maintenance-intensive, single-user and error-prone.

• Implementation: Migrated all functionality and VBA macros into a standalone web app with central data storage, multi-user support and automated reporting.

• Cash-flow modelling

• Starting point: The existing cash-flow model covered standing investments only; project developments were missing from steering.

• Implementation: Built and extended the CF model to include project-development cash flows.

• Optimisation: Reviewed and optimised existing CF models and expanded the KPI outputs for reporting and steering.

• Used OpenCode and AI-assisted software engineering to design, implement, refactor, test, and document an end-to-end RAW/SDTM/ADaM pipeline in R for Dr. Falk Pharma (07/2026), including metadata-driven transformations, automated validation rules and QC, traceability, and reproducible clinical outputs.

• Design and build of a fully isolated AWS offline environment with no outbound internet access for running a browser-based business application
• Design and implementation of a proxy and response service that terminates all external application calls inside the VPC and serves them from locally stored content; identification of the actual communication needs through measurement-based DNS query logging
• Creation of architecture designs and decision papers including a comparison of options (Application Load Balancer with Lambda and S3, reverse proxy on EC2, private API Gateway) assessed by operational effort, cost, and availability
• Transfer of the solution and operations documentation previously available only for Azure to an AWS target architecture, including reassignment of all services and operational processes
• Automated rollout as Infrastructure as Code (Terraform, CloudFormation) with CI deployment via GitHub Actions, plus setup of private DNS zones and an internal certificate chain for operation without internet access
• Creation of architecture, deployment, and operations documentation and handover to the customer
• Build-up of a private cloud platform on OpenStack at provider TelemaxX with Terraform, including FortiGate HA clusters, FortiManager, and Kubernetes
• Introduction of Policy as Code (Open Policy Agent, Conftest) as well as development of MCP servers (Model Context Protocol) to connect AI assistants to operations and project tools

Successes:

• Made the business application fully operable without internet access for the first time; the cause of the loading error was narrowed down systematically to missing CORS headers after the likely certificate issue was ruled out
• Fully transferred an existing Azure concept to AWS and replaced the manually created environment with a reproducible, CI-based rollout

Technology stack: AWS (VPC, Application Load Balancer, Lambda, S3, Route 53 private hosted zones and Resolver query logging, IAM, CloudWatch, EC2, CloudFormation), Infrastructure as Code (Terraform, CloudFormation, Remote State), CI/CD (GitHub Actions with OIDC, Azure DevOps Pipelines), OpenStack, FortiGate, FortiManager, Kubernetes, Policy as Code (Open Policy Agent, Conftest), offline and air-gap architectures, PKI & certificates (internal CA, TLS, CRL/OCSP), DNS, network segmentation, Linux, Windows Server, Python, Bash, PowerShell, YAML, JSON, architecture design & decision papers, documentation (Confluence, Markdown), Generative & Agentic AI (Model Context Protocol, Agentic AI Coding Tools)

• New development and further development of frontend interfaces for investors, using HTML5, JavaScript, CSS, React, Material UI, and JSTL

• PRISMA project
• Ansible automation
• Windows third-level support for Windows NT, Windows 2000, Windows 2013, Windows 2016, Windows 2019

• Working on Telecommunication Apps on Android under NDA for the Military
• Participation in the development of backend services (Spring, Java)
• Android Development, Kotlin, Compose, REST, JSON, State-Machines, Android Studio, IP Communication, Cybersecurity, Agile

• Operations and extensions of Azure DevOps pipelines

• Operations and extensions of AWS services

• Citrix (Windows 10, Bitwarden)

• AWS: ECR, EKS, CloudFront CDN, Route 53, VPC peering and CNI upgrade, Atlas MongoDB, S3 buckets, static website hosting

• Azure: build and deploy with DevOps pipelines

• LLM Benchmarking: Evaluated and benchmarked 1,000+ LLM-generated responses for technical correctness, code efficiency, safety, and strict adherence to complex technical prompt constraints.
• Mathematical Auditing: Executed highly granular structural audits of model-generated code and multi-step mathematical derivations, documenting root-cause logic gaps, syntax errors, and hallucinations.
• Prompt Architecture: Authored and stress-tested high-complexity mathematical, logical, and coding prompts designed to uncover edge-case failures and assess advanced reasoning capabilities.
• Model Alignment: Led RLHF and SFT workflows, utilizing Google Collab and interactive Python scripts to parse massive datasets of model outputs and deliver highly detailed preference rankings.

Two complementary tracks: (a) building my own AI products and open-source tools under the Neckarshore AI brand, with a focus on AI-supported multi-agent development processes and documentation automation; (b) freelance support for industrial, logistics, and financial companies in IT-related transformation projects.

Current product portfolio under the Neckarshore AI brand:

• Omnopsis Documenter — AI-powered documentation engine: generates compliance, technical, and release documentation from Git, Jira, and Confluence. 500+ automated tests, 100+ API endpoints, RBAC, monitoring stack.
• TrustScope — trust analysis for public GitHub repos: deterministic three-pillar report (Security & Supply Chain, Governance, Community) based on OpenSSF Scorecard, with constructive fix suggestions instead of a single misleading score.
• md-viewer — dependency-free Markdown viewer for macOS (Finder Quick Action) and web, split view of rendered document and raw text.
• Obsidian Vault Autopilot — AI-supported vault automation for Obsidian × Claude Code: sorts the inbox, renames notes, enriches frontmatter.
• Phonesis Voice Bank — voice archive for families, hospices, and cultural archives (guided recordings, inheritable archive, optional voice cloning); currently in the GDPR compliance phase (Art. 9, biometric data).
• Multi-Agent Development Process — 12+ specialized agents (architecture, implementation, security, refactoring, marketing) with structured handoff protocol, parallel execution, and automated quality assurance — powers the entire product development.
• Freelance consulting: Technical Product Ownership for production, logistics, and compliance/AML systems; steering international rollouts including test, cutover, and change management.

Stack: NestJS, TypeScript, PostgreSQL, Redis/BullMQ, Next.js, Claude Code (Opus), Docker, GitHub Actions, Vercel

Worked as an App Engineer as part of the IoT Control team in a React Native Stack.

• Implementing new Components based on the Tonies Design System.
• Improving, extending and maintaining Tonies Design System
• Adding new features like Volume and Playback Control of the Tonies
• Fixed connection issues via bluetooth and wifi
• Removing technical debt and polishing for an upcoming big software and hardware release
• Improving accessibility by optimizing screens and components for usage via screen readers.
• Wrote lots of automated tests to ensure app quality is always stellar.
• Weekly app releases in a high volume workload environment

Results & Impact:

• Improved app stability and connection reliability ahead of a major software and hardware release
• Raised overall app quality through a significantly broader automated test suite, catching regressions before they reached users
• Made the app usable with screen readers, improving accessibility for a wider audience
• Sustained a weekly release rhythm in a high-volume production environment without quality regressions

Technologies: react-native, expo, typescript, react-19, react-compiler, react-navigation, zustand, @tanstack/react-query, i18next, figma, app-store, play-store, sentry, bitrise, config-cat, poeditor