Projekte Case Studies

Ausgelieferte KI-Automatisierungs-Projekte für DACH-Unternehmen. Problem, Lösung, Metriken, Links zum Code. Shipped AI automation projects for DACH companies. Problem, solution, metrics, links to the code.

Context Layer for AI Agents: Sourced Answers and Human-Approved Actions Across Chat, Tickets, Wiki and CRM
▶︎ 0:47 Governed context workflow

Context Layer for AI Agents: Sourced Answers and Human-Approved Actions Across Chat, Tickets, Wiki and CRM

September 5, 2026 · 14 min read
Case study: a context layer for AI agents on a temporal knowledge graph. Sourced answers from chat, tickets, wiki and CRM, external actions authorized before execution.
Production Self-Hosted Voice AI Platform For Data-Residency-Sensitive Teams
▶︎ 1:09 live console demo inside

Production Self-Hosted Voice AI Platform For Data-Residency-Sensitive Teams

May 21, 2026 · 4 min read
GPU-backed self-hosted voice AI on Dograh: local STT, LLM, TTS in EU infra or client VPC. Measured warm latencies, structured outcome writeback.
AI Revenue Prioritization System Embedded in HubSpot CRM

AI Revenue Prioritization System Embedded in HubSpot CRM

April 23, 2026 · 5 min read
HubSpot-native AI layer that scores every contact on ICP fit, ranks the pipeline by expected revenue, and prescribes a specific next action with a written reasoning trace — all as first-class HubSpot properties, no tool switching.
AI Decision Support Platform for Enterprise Operations

AI Decision Support Platform for Enterprise Operations

April 22, 2026 · 4 min read
Transformation portfolio platform with AI-scored initiatives, capability-maturity radar, and investment recommendations. Replaces steering-committee spreadsheets with continuous prioritization across business value, complexity, maturity, and ROI.
Enterprise AI PII Redaction System for Sensitive Documents
▶︎ 1:43 live console demo inside

Enterprise AI PII Redaction System for Sensitive Documents

April 21, 2026 · 6 min read
Self-hosted German PII redactor for SAP prod→dev copies. Plugs in after TDMS/Delphix/Informatica to cover free-text NOTES columns, unclassified Z-tables, and OCR'd attachments. DSGVO-konform, Apache 2.0, runs on a single consumer GPU.
The context layer for your AI agents

Your agents answer from whatever the retriever finds, and too often that is last quarter's truth. I build the context layer they answer and act from: a temporal knowledge graph that keeps every fact with its source and the time it held, reads with each person's own permissions, and writes nothing without a person's approval. On your own tenant, billed by the hour, step by step.