Dennis Huber. DE LinkedIn

AI Engineer · Tübingen, Germany

Dennis Huber

I take AI from prototype to production.

At SSC-Services I build AI solutions for industrial customers: models that understand engineering drawings, automate manual review processes and integrate directly into CAD systems like Siemens NX. In the EU-funded FFplus programme I built LLM4CAD together with the Karlsruhe Institute of Technology.

Contact me on LinkedIn See projects

  • Open to AI engineering roles
  • Stuttgart / Tübingen area, hybrid
Portrait of Dennis Huber

Projects

Selected work

EU-funded · FFplus

LLM4CAD: extracting information from engineering drawings

Many companies store their 2D engineering drawings only as images. Key information in title blocks, tables and dimensions is hard to search, and conventional text recognition fails on dense layouts.

We trained vision-language models on synthetically generated drawings to read this information reliably. Training ran on European high-performance computing infrastructure, while the model runs on-premises so sensitive engineering data stays in-house. Our technology partner was KIT.

  • Vision-language models
  • Fine-tuning
  • Synthetic data
  • PyTorch
  • HPC
less time to extract key metadata
−92 %
less manual effort
−96 %
drawings tested
2,600
GPU node hours of training
6,500

Source: FFplus success story

Automated IP-rights recognition

For a strategic key customer I developed an object detection model that takes over a previously manual review process and automates the downstream steps. It runs in production on Azure and laid the groundwork for LLM4CAD.

  • PyTorch
  • Object detection
  • Azure

AI applications with clear guardrails

LLM-based applications that combine templates, data and tools to take repetitive work off business teams, with defined rules for which actions the model may take, how errors are handled and when a human decides.

  • LLMs
  • Agents
  • Guardrails
  • Python

Add-ons for Siemens NX

Native extensions and automation in C# that bring AI and process logic straight into the CAD environment engineers already work in.

  • C#
  • NX Open
  • CAD integration

Own server infrastructure

In my spare time I run my own Linux server with containerised applications, including a web app I wrote in Rust. This website also runs on my own infrastructure.

  • Linux
  • Containers
  • Rust

Experience

Experience & education

  1. since 04/2024

    Software Developer / AI Specialist

    SSC-Services GmbH, Böblingen

    • Design and development of ML and object detection models (PyTorch) for image and document analysis
    • LLM4CAD in the EU funding programme FFplus, together with KIT
    • Native NX add-ons and automation tools in C#
    • Backend and automation solutions in Python and Rust on Linux
    • Architecture designs and technical consulting for internal and external stakeholders
    • Designed and delivered the internal “AI Basics” training, now available as a webinar
  2. 03/2023 – 03/2024

    Working Student

    SSC-Services GmbH

    • CNNs for object detection in image analysis
    • Web-based email automation tool
    • Data conversion scripts and software documentation
  3. 04/2017 – 04/2024

    B.Sc. Computer Science

    University of Tübingen

  4. 2014 – 2017

    Electrical Engineering & Information Technology

    University of Stuttgart · Esslingen University

Skills

Artificial intelligence

  • PyTorch
  • Vision-language models
  • Object detection
  • LLMs & agents
  • Fine-tuning

Development

  • Python
  • Rust
  • C#
  • Siemens NX Open

Operations

  • Azure
  • Linux
  • Docker / Podman

Languages

  • Germannative
  • Englishfluent
  • Spanishgood working knowledge

Training

  • AI training under the EU AI Act Smart Data Solution Center · 2025
  • AI models, systems and literacy Smart Data Solution Center · 2025
  • Agile requirements engineering oose eG · 2024
  • Conflict management oose eG · 2024

Talks

Sharing knowledge

  1. AI in engineering design: new possibilities in practice

    NXT nAIght, Reutlingen University (in German) · event

  2. LLM-based CAD Information Extraction

    SSC Summerfest AI 2.0, Stuttgart · presenting the FFplus study · report

  3. AI in engineering design

    Expert talk at VDI, the Association of German Engineers

  4. AI Basics

    Training for colleagues, designed and delivered, later recorded as a webinar

Let’s talk.

I’m happy to hear about AI engineering roles in the Stuttgart and Tübingen area. I’ll gladly send you my CV and references.

Message me on LinkedIn