AI in everything I do·Munich·EN / DE

Linus Teklenburg.

I turn research into deployed AI systems - and AI governance into working software. Currently shipping Multi-Agent RAG with MCP at a Big 4 company.

Get in touch LinkedIn
Scroll
● Currently · Since Jan 2026

Assistant Manager &
AI Ambassador at a Big 4 company.

Double-promoted from Associate - skipping a level in twelve months. Building Multi-Agent RAG systems with MCP (Model Context Protocol) for context-aware enterprise automation - and leading firm-wide Trusted AI enablement across finance, automotive and public sector.

MCP RAG Multi-Agent Systems ServiceNow Azure OpenAI EU AI Act Trusted AI
MCP
Trusted AI
EU AI Act
GenAI
Multi-Agent
ServiceNow
XAI
Azure OpenAI
0
Years programming
0
Products shipped
1.0
Published thesis grade
0
Years founding
0
Published & speaking
The Journey

From first websites
to enterprise AI delivery.

Nine years of shipping - from weekend freelance HTML at 14, through a first-class AI degree, two company foundations, and now engineering trusted AI for the Big 4.

2017 - 2020

Full-Stack Web Development

Freelance · consult.IN e.V. · Ingolstadt

Started shipping websites at 14 for local SMEs - restaurants, craftsmen, small e-commerce. Learned the craft end-to-end: turning a client conversation into a live product. Later joined consult.IN e.V. (Oct 2020 - Feb 2022), a student consulting initiative in Ingolstadt, where client-facing delivery became second nature.

  • Built and shipped 15+ websites and landing pages for SMEs
  • Custom WordPress themes, Stripe & PayPal payment integrations
  • Responsive, SEO-optimised, Lighthouse > 90 on launch
  • End-to-end delivery: discovery → design → dev → deploy → maintain
HTML5 CSS3 JavaScript PHP WordPress SEO Stripe
2020 - 2024

B.Sc. Artificial Intelligence

Technische Hochschule Ingolstadt · 2020 - 2024

Focused on Computer Vision, Transformers and Reinforcement Learning - with a strong foundation in classical ML and math. Bachelor thesis combined YOLOv8 real-time object detection with XGBoost to predict American football plays from broadcast video - published open-access with grade 1.0 on opus4.kobv.de.

  • Classical ML: Supervised & Unsupervised Learning, SGD, SVM, Random Forest, Decision Trees, k-NN, Naive Bayes, Clustering, PCA, Linear & Logistic Regression
  • Deep Learning: CNNs, RNNs, LSTMs, Transformers, Attention, Transfer Learning, Fine-tuning
  • Core areas: Computer Vision, NLP, Reinforcement Learning, Bayesian methods
  • Math foundations: Linear Algebra, Multivariate Calculus, Probability, Statistics, Optimization
  • Software Engineering: Algorithms, Data Structures, OOP, Design Patterns, Architecture
  • Thesis: end-to-end pipeline from video ingestion → player tracking → play prediction
Python PyTorch TensorFlow Scikit-learn OpenCV YOLOv8 XGBoost Random Forest SVM SGD k-NN PCA Clustering Transformers Hugging Face NLTK · spaCy Reinforcement Learning CNNs · RNNs · LSTMs NumPy · Pandas Jupyter LaTeX
Published thesis (PDF)
Aug 2021 - Dec 2025

Co-Founder · Stratek GmbH & GbR

Wolnzach · 4½ years · 10+ products shipped

Co-founded Stratek while studying - grew from a two-person GbR to a full GmbH. Shipped 10+ web & mobile products and engineered a computer-vision drone pipeline for aerial inspection.

  • 10+ Flutter & web apps live on App Store and Google Play
  • Drone + Computer Vision pipeline: real-time detection on edge hardware
  • AI-integrated products: RAG chatbots, LLM document processing, vector search
  • Full product ownership - discovery, design, architecture, delivery, support
  • Agile / Scrum cadence, OKRs, design thinking workshops with clients
Flutter React Node.js TypeScript Python Firebase AWS Docker YOLOv8 RAG OpenAI API Pinecone Scrum Design Thinking
Dec 2024 - now

Associate → Assistant Manager + AI Ambassador

Big 4 consulting · Munich · promoted Jan 2026

Joined as Associate. Twelve months later, double-promoted (two levels in one jump) into two parallel roles.

  • Assistant Manager - Design and deploy AI agents on ServiceNow. Develop our Multi-Agent RAG system using MCP (Model Context Protocol) for context-aware enterprise automation.
  • AI Ambassador - Lead firm-wide workshops on Generative AI, EU AI Act compliance, and Trusted AI. Coordinate the internal AI community and bridge strategy, prototyping, and business value across finance, automotive & public sector.
  • ServiceNow Certified System Administrator (Sept 2025)
MCP RAG LangChain AutoGen ServiceNow Copilot Studio Azure OpenAI Semantic Kernel EU AI Act Trusted AI XAI Bias Detection Workshop Facilitation
Technical Stack

From models to production.

Full-stack across the modern AI delivery pipeline - research, agents, cloud, governance, and the consulting craft on top.

🧠
AI / ML Core
Python PyTorch TensorFlow OpenCV YOLOv8 Transformers Hugging Face NLTK spaCy Scikit-learn XGBoost NumPy · Pandas Reinforcement Learning
🤖
Agentic AI & LLMs
MCP (Model Context Protocol) RAG Multi-Agent Systems AutoGen LangChain LangGraph Semantic Kernel OpenAI API Azure OpenAI Pinecone · FAISS Fine-tuning Prompt Engineering
🛡️
AI Governance
EU AI Act Trusted AI XAI (Explainable AI) Bias Detection AI Risk Management Model Cards ISO 42001 NIST AI RMF Ethical AI
☁️
Cloud & Platform
Azure OpenAI Azure AI Foundry ServiceNow Copilot Studio AWS (EC2 · S3 · Lambda) Firebase Docker GitHub Actions · CI/CD Power Platform
⚡
Full-Stack Engineering
React · Next.js Flutter · Dart Node.js · Express TypeScript FastAPI PostgreSQL · MongoDB GraphQL · REST WebSockets Tailwind
🧭
Consulting Craft
Stakeholder Engagement Workshop Facilitation Pre-Sales Support Proposal Writing Client Presentations Change Management Scrum · Agile OKRs Design Thinking
Open Source

Code in public.

Weekly experiments on MCP, agents, and whatever I'm curious about this week. Live pulled from github.com/linus131313.

—
Public repos
—
Last commit
linus131313/experiments
Open on GitHub
Recent repositories
loading repo

loading description text here while we fetch data

loading
loading repo

loading description text here while we fetch data

loading
loading repo

loading description text here while we fetch data

loading
loading repo

loading description text here while we fetch data

loading
Syncing from GitHub …
View all on GitHub →
Research & Talks

Published. Spoken. Certified.

I believe AI in the enterprise has to be trusted, explainable, and actually deployed - not just demoed.

Independent Research · 2026 · Narrative review

Reliability Patterns for LLM-Based Multi-Agent Systems

A Synthesis of Published Evidence.

A review of six engineering patterns commonly invoked to make language-model agents reliable - structured tool schemas, validator agents, reflection-and-retry, least-privilege tool scoping, failure propagation, and deterministic output formatting. Each pattern is mapped to the strongest controlled evidence in the literature and graded on a strong / moderate / weak / absent rubric.

  • Three patterns well-supported by independent controlled studies
  • Three widely recommended but resting on thinner isolation ablations
  • Cross-cutting finding: reliability depends on an external grounding signal - pure self-assessment loops are fragile
  • Five open research questions identified for the field
Read the paper (PDF)
Peer-reviewed · Open access · 2024

YOLOv8 + XGBoost for NFL play prediction

Bachelor thesis combining real-time object detection with gradient-boosted decision trees to predict American football plays from broadcast video. Grade 1.0, published open-access.

opus4.kobv.de
Independent Research · 2026 · Pre-print

The MCP Governance Gap

Empirical Evidence on Dynamic Tool Binding in Enterprise AI.

A quantitative study of the gap between Model Context Protocol adoption velocity and the response cadence of ISO/IEC 42001, NIST AI RMF, and the EU AI Act. Introduces the Tool Graph Capability Score (TCS) - a reproducible metric that places a runtime agent on the same axis as traditional approved applications.

  • Sample of n = 200 MCP-server repos from a population of ~98,000
  • 10 public CVEs mapped against 6 control breakpoints
  • A typical full-stack MCP configuration expresses 7.1× the capability surface of an approved DBA-plus-shell tool
  • Robust to 48 reweightings in sensitivity analysis
Read the paper (PDF) Code on GitHub

Munich Datageeks e.V.

Two-time guest speaker · 2024 - now
  • Transformer architectures & enterprise applications
  • Agentic AI · Multi-Agent Systems · Computer Vision in industry
Talk announcement on LinkedIn

Certifications

Professional credentials
  • ServiceNow Certified System Administrator · Sept 2025
  • EU-Z Basiszertifikat
  • EU-Z Vertiefungszertifikat

Ongoing research interests

What I'm working toward
  • Trusted AI frameworks for regulated industries
  • Multi-agent orchestration patterns with MCP
  • Evaluation & guardrails for production LLM systems
  • Cost-efficient RAG architectures at scale
Beyond Work

Leadership on the field.
Awards off it.

What I do when I'm not shipping code.

American Football · Ingolstadt Dukes

2018 - 2024 · 6 years · Leadership & coaching
  • 2× Seniors Team Captain + Starting Linebacker
  • 2× U19 Team Captain
  • 1× U19 Defense Coordinator (coaching role)

Six years of competitive team sport - leading the locker room, calling the defense, coaching the next generation. The best management training you can get in your twenties.

dukesingolstadt.de
1st Place · Sustainability Prize

BBW e.V. Startup Competition

Built "StudyMatch" - a Flutter app for matching students by study goals - end-to-end in 4 weeks. Design, development, App Store + Play Store deployment. Won 1st Place and the Sustainability Award at the BBW Gründerpreis.

  • Full Flutter build: design → code → release in 4 weeks
  • Published to both App Store and Google Play
  • 1st place overall + Sustainability Prize
Press release on thi.de Press release on donaukurier.de

Let's talk. 👋

Drop a line. I read every message.

Message sent.

Thanks for reaching out - I'll reply within 48 hours.

or directly: hey@linus-teklenburg.de LinkedIn GitHub
LinkedIn GitHub 📧 Contact