Cases

What we build and who we advise — practice, not PowerPoint.

Own developments

Our own AI systems as proof: we know what we're talking about.

Mingly — Multi-LLM Desktop App

Mingly brings multiple AI models (Claude, GPT, local Ollama models) into a native macOS desktop app. In production since early 2026, with focus on privacy and multi-model routing.

ElectronReact 19TypeScriptTailwind

DocMind — RAG Frontend for Private Data

DocMind is a desktop application that makes private document collections (PDFs, notes, wikis) searchable — fully offline, with locally running embeddings and Qdrant as vector store.

ElectronReactTypeScriptQdrant

RAG-Wissen — Production RAG for our own knowledge base

RAG-Wissen is our internal RAG system for advisory knowledge, technical documentation, and market intelligence. Runs in production with Qdrant v1.17.1 and provides the research base for our blog articles.

Python 3.13QdrantFastAPIMCP

Prüfstand — AI Testing Framework

Prüfstand is our framework for systematic testing of AI systems — prompt variations, model comparisons, quality scoring via LLM-as-a-Judge. Used for quality assurance of our own products.

Electron GUIPython BackendVitestPytest

Nexbid — Agentic Ad Server

Nexbid is a platform for agent commerce: advertising that AI agents can understand, evaluate, and transact with. Live with x402 payment pilot, MCP server, and AdCP 3.0 trust surface.

TypeScriptNode.jsVercel FunctionsNeon Postgres

Compass — Research Project

Compass is an internal research project investigating agentic workflows in complex research tasks.

PythonClaude Agent SDKMCP

Eval-Framework — LLM-as-a-Judge

Our framework for systematic evaluation of LLM outputs — with pairwise comparisons, bias corrections, and calibration against human baselines.

Python 3.13SQLitePydanticPrometheus2-Judge