About

I'm Christian Nicolussi (geb. Bernhard) from Munich.

I lead the AI team at CANCOM. My work sits at the intersection of GenAI engineering, hardware-aware system design and translating all of that into language a board actually understands.

Most of my days look like one of three things: designing and shipping production-grade LLM, RAG and agentic systems across cloud and on-prem; advising C-level stakeholders on cost-optimised, sovereign AI strategies; or running workshops that move teams from “we should do something with AI” to a working PoC. The timeline below turns each station of my career into its own case study.

Christian Nicolussi (geb. Bernhard) on stage
Download CV
5+
Years in software & AI
100+
Workshops, talks & trainings delivered
M.Sc.
Computer Science, LMU Munich
20+
AI certifications

Skills

Build & ship AI applications

Voice AI agentsRAG & knowledge systemsMultimodal / VLMMulti-agent workflowsRealtime voice (LiveKit · Realtime & STT/LLM/TTS pipeline)Demos, PoCs & product deliveryNext.js · React · FastAPI

Models, agents & quality

Open & closed LLMs (GPT, LLaMA, Qwen, Nemotron, …)Fine-tuning (LoRA / QLoRA)Embedding modelsLangGraph · LangChainContext engineeringEval & observability (RAGAS, Langfuse)MCP / tool use

NVIDIA & production inference

NVIDIA AI EnterpriseNIMsTensorRT-LLMvLLMNemotronDGX Spark · H100 · H200 · DGX B300GPU sizing & TCOPyTorch · HuggingFace

Cloud, platform & delivery

Azure AI (Foundry · OpenAI · Search)Sovereign & on-prem AIDocker · KubernetesPython · TypeScriptEnterprise & regulated deployments

Lead & enable

Technical workshops & keynotesC-level & executive advisoryTeam leadership (FDE teams)Pre-sales → production delivery

Education

  • M.Sc. Computer Science

    Ludwig-Maximilians-Universität München · 2022 — 2024

    Thesis at BMW: "Fine-Tuning Retriever and Generator Components in RAG — a Cross-Domain Study" (1.4).

  • B.Sc. Business Informatics

    University of Applied Sciences Munich · 2018 — 2022

    Bachelor thesis: "Natural Language Processing — Effort Estimation in Agile Projects" (1.8).

Always learning

View all →
  • 20+ AI certifications across NVIDIA Deep Learning Institute, DeepLearning.AI and Coursera
  • Covering LLMs, RAG, Agents, Evaluation, LangChain, LangGraph and MCP
  • HackaTUM 2023 (TU Munich) · Kaggle: HuBMAP — Hacking the Human Vasculature
  • LLaMA-2 implemented from scratch in PyTorch for research

Languages

  • GermanNative
  • EnglishFluent
  • SpanishBeginner
  • MandarinBeginner

Get in touch