LiveAI Products
AIUseCaseHub
A public catalog of more than 4,000 real enterprise AI deployments, each traced to a public source, with insights and trend analysis on top.
4,015source-linked AI deployments
Senior Cloud Solution Architect at Microsoft
Florian Follonier
I turn AI strategy into secure, operable systems, and I build my own products and open-source experiments to find out what actually holds up.
I work on applied AI: agents, evaluation, and the engineering that takes a prototype into production. Outside customer work I build products and open-source experiments, and write about what I learn.
Zurich area, Switzerland
PhD, Univ. of Bremen
- enterprise AI deployments cataloged
- 4,015
- live products I run
- 4
- projects with public code
- 7
- research publications
- 7
Selected work
Things I've built and run
Live products, open-source tools, and research prototypes, each built to answer a real question.
Live
Open sourceEvaluation
Swiss German Voice Benchmark
A local-first workbench for measuring how well voice models understand Swiss German dialects, with WER, CER, chrF, and streaming latency.
200SwissDial clips, 25 per dialect
In progressAgents
Milo: Embodied Robot Lab
An in-progress robotics research simulator for embodied agents, with a React operator cockpit, agent-guided navigation, sensor telemetry, and reproducible test scenarios.
20+reproducible test scenarios
Open sourceData Pipelines
AI Web Scraping Pipeline
An open-source agentic pipeline for discovering, extracting, deduplicating, and reviewing structured evidence from public websites. It powers AIUseCaseHub.
3agents: discovery, extraction, review
Open sourceAgents
RFP Agent on Microsoft Foundry
A Microsoft Agent Framework and Foundry demo that turns a customer opportunity into a cited, enterprise-ready solution proposal, visualized as a pixel-art agent office.
5progressive demos, from baseline to observability
LiveWriting
Recent notes and results
Project Update
What I have been building since the summer
Two public market tools, a Swiss German speech benchmark, an MCP server, a robotics simulator, and a few open-source evaluation projects.
Project Update
How well do voice models understand Swiss German?
First results from my Swiss German benchmark: 200 SwissDial clips across eight dialects, comparing OpenAI Realtime 2 and Realtime 2.1.
Project Update
AIUseCaseHub is now available to agents through MCP
A hosted MCP server with four tools for searching source-linked enterprise AI deployments, with a free public preview.
Now
Currently exploring
Benchmarking voice models on Swiss German dialects. Across 200 SwissDial clips, Realtime 2.1 led Realtime 2 on both High German translation and verbatim dialect transcription.
Building Milo, a robotics simulator for embodied agents, where a local model makes navigation decisions and a larger model supervises tasks within explicit token and request budgets.
Opening AIUseCaseHub to agents with a hosted MCP server, and growing the record past 4,000 source-linked enterprise AI deployments.
Shipping public market-research tools: Rankoon for ML-ranked crypto signals with published hit rates, and PerpXray for read-only Hyperliquid trading analytics.
Research
Selected publications
My research at the University of Bremen explored service innovation, modularity, and collaboration in automotive engineering.
Patterns of Service Innovation: An Empirical Study of Automotive Engineering Services in Germany
University of Bremen · Doctoral dissertation
Service Innovation: Sensing with Information Systems
Americas Conference on Information Systems (AMCIS) · Conference paper
Schools of Thought in Service Modularity
European Conference on Information Systems (ECIS) · Conference paper
Speaking & collaboration
Let’s talk about your AI project.
I give talks and workshops on AI agents, evaluation, and cloud architecture, and I’m happy to discuss a session, a collaboration, or a project.