Skip to content

Experience

Where I've worked and what I've built.

BCG Platinion, SAP, Allianz, Fraunhofer, and TUM. What I've worked on and what I've learned along the way.

Track Record

Timeline

Product Owner & AI Architecture Lead (Internship)

SAP

April 2026 – July 2026·Munich, Germany·On-site

Product owner and architecture lead for Mycellium, an orchestrator-free multi-agent system built as part of the TUM SAP Practical Course. Designed the full system architecture and led a team of 6 through research, implementation, and stakeholder presentations.

  • Orchestrator-free agent swarm architecture with capability- and trust-weighted bidding, an append-only blackboard, and consensus built through agents citing and building on each other's findings
  • Contract and thread decomposition, the auction protocol for self-selection, and the blackboard tool suite (post, cite, vote, mention)
  • UI: visual canvas playground for configuring agent networks and simulating live swarm runs with step-by-step playback
  • Product ownership: backlog management, sprint coordination, and technical presentations to SAP industry stakeholders

Working Student, AI & Data Analytics

BCG Platinion (Boston Consulting Group)

November 2025 – Present·Munich, Germany·On-site

Contributing to AI strategy initiatives within the Energy practice. Co-designing stakeholder workshops on AI adoption. Delivering AI-powered solutions for enterprise clients.

  • AI strategy for energy value chains: GenAI adoption, architecture design, feasibility
  • Stakeholder workshops on AI impact, risks, and adoption strategies
  • Technical architecture slides: GenAI system design, data pipelines, governance

Working Student, AI/ML

Allianz SE

October 2024 – November 2025·Munich, Germany·Hybrid

Developed GenAI applications for insurance operations. Designed agentic workflows with cost-optimized model selection. Drove adoption through stakeholder workshops.

  • GenAI applications: customer communication analysis, medical report insights, insurance evaluations
  • Agentic workflow design with cost-benefit analysis across models and cloud providers
  • Prompt engineering framework presented in stakeholder workshops
  • CI/CD pipelines (Jenkins) for promoting agentic workflows to production

Intern, Generative AI

Fraunhofer Society

March 2024 – July 2024·Munich, Germany·Hybrid

Built a private RAG pipeline for institutional knowledge management. Led improvements in data ingestion, retrieval quality, and AI security.

  • Private RAG pipeline using institutional public data
  • Data ingestion optimization and retrieval quality improvements
  • Guardrail design: anti-hallucination, prompt attack resistance
  • Prompt engineering for reliable, secure knowledge management

Research Author, LLMs for Science

Published in JSEP (Journal of Software: Evolution and Process)

October 2023 – January 2024·Munich, Germany·Research

Co-authored a peer-reviewed paper on the use of LLMs in research and data analysis. The work has been cited over 200 times (70+ on Scopus), contributing to the academic discourse on AI adoption.

  • Peer-reviewed publication on LLM usage for code generation and data analysis
  • 200+ citations, 70+ on Scopus, serving as a foundation for AI-driven research training
  • Contributed to the academic discourse on responsible AI adoption in scientific workflows

Skill map

Capability Graph

Search by skill, company, project, or technology and explore how capabilities connect.

Showing 13 of 13 nodes.

Technical Skills

Search across the full skill set, then scan only the categories that match.

Showing 66 of 66 skills across 8 categories.

AI & Machine Learning

core
Large Language Models (GPT, Claude, Gemini, Bedrock Models)Agentic AI SystemsMulti-Agent ArchitectureRetrieval-Augmented Generation (RAG)Hybrid Search & RetrievalPrompt Engineering & OptimizationAgentic Workflows & OrchestrationFlow Engineering (Bedrock Flows)Document Intelligence & NLPMulti-Model OrchestrationHallucination MitigationAI Security & GuardrailsEvaluation & Reliability for LLM Systems

Frontend Engineering

core
TypeScript / JavaScriptNext.js / ReactTailwind CSSResponsive UI SystemsAccessibility (WCAG)Component Architecture (Radix UI)Data Visualization for AI Workflows

Backend Engineering

core
PythonJavaC++ / SystemCSQLFastAPIJava Spring BootNode.jsREST APIs & MicroservicesPostgreSQL / Vector DatabasesOCR & Document Processing Pipelines

Cloud & DevOps

working
AWS (Bedrock, SageMaker)Azure / Azure OpenAIVercel DeploymentDockerCI/CD (Jenkins)

Architecture & Product

working
Solution ArchitectureSystem Design PatternsAgent Protocol DesignData Pipeline DesignProduct Ownership for AI Projects

Strategy & Delivery

working
AI Feasibility AssessmentCost-Benefit AnalysisModel / Provider Tradeoff AnalysisStakeholder Workshop FacilitationAgile / Scrum (Jira, Confluence)

Research & Foundations

working
Trustworthy AI ResearchPrompt-Time Risk AnalysisEmpirical Evaluation DesignScientific Writing & PublicationComputer Architecture (FPU Design)

Ecosystem & Tooling

working
LangChain / FlowiseClaude Agent SDKOpenAI APIsGemini APIsPrompt Frameworks for Enterprise Use Cases

Capabilities

Languages and domains

Languages

  • English (Fluent)
  • German (Fluent)
  • French (Native)
  • Arabic (Native)
  • Spanish (Basic)

Sectors

  • Enterprise AI / Consulting
  • Insurance & Financial Services
  • Energy & Utilities
  • Research & Higher Education

Ways of Working

  • Agile / Scrum delivery
  • Stakeholder workshop facilitation
  • Feasibility & cost-benefit analysis
  • Architecture & solution design

This website can use privacy-friendly analytics to improve content quality. No advertising cookies are used.

You can change this anytime in the privacy page. Read more in Privacy Policy.