I’m David Graf, a Swiss software engineer with a passion for turning complex problems into elegant solutions. With 6+ years of professional software development experience and 3+ years specializing in LLM-powered knowledge systems, I bridge the gap between cutting-edge AI research and practical applications that customers actually want.
What Drives Me#
Unlike many engineers who prefer staying in their technical comfort zone, I thrive on diving deep with users and customers to truly understand their challenges. I believe the best solutions come from this intersection of technical expertise and real-world problem understanding. I’m always ready to learn whatever it takes to deliver results that matter.
My Expertise#
I specialize in building AI-powered knowledge management systems that extract meaningful insights from complex, unstructured data. My work focuses on:
- Knowledge Graphs & Graph Neural Networks - Building intelligent systems that understand relationships and connections in data
- Large Language Models - Integrating LLMs into production systems for real-world applications
- Semantic Search & Recommendation Systems - Creating advanced search solutions that deliver highly relevant results
- Research & Trend Analysis - Developing ML systems that identify emerging patterns and opportunities
- Production ML Systems - Taking prototypes to scalable, robust production deployments
Recent Highlight#
I built the Research Topic Launcher for a leading scientific publisher - an end-to-end ML system that revolutionized how they discover research trends and identify experts among 200+ million papers. The system reduced discovery time from weeks to minutes, with over 90% user preference rate compared to previous manual methods.
The project combined custom knowledge graphs, embedding models, and LLM-powered analysis to create capabilities that didn’t exist before. It’s exactly the kind of challenge I love: no clear existing solution, requiring innovation under uncertainty, and delivering measurable business value.
Technical Foundation#
My technical toolkit spans the full ML lifecycle:
- Machine Learning: PyTorch, scikit-learn, embedding models, recommendation systems
- Knowledge Graphs: Neo4j, NetworkX, graph algorithms, dynamic graph generation
- LLM Integration: OpenAI API, prompt engineering, evaluation frameworks
- Data Engineering: Google BigQuery, Dagster, large-scale data processing
- MLOps: Azure ML Studio, MLFlow, Docker, Terraform, CI/CD pipelines
Background#
I hold a Master’s in Computer Science from ETH Zürich, where I specialized in knowledge graphs and graph neural networks. My thesis work was conducted in collaboration with IBM Research, giving me early exposure to industry-academic collaboration.
I’m multilingual (Italian, English, German, French, and some Macedonian) and have experience working with international teams. I completed my military service as an officer in the Swiss Army, which taught me valuable lessons about leadership and working under pressure.
What I’m Looking For#
I’m currently seeking opportunities where I can apply my expertise in LLMs and knowledge graphs to solve meaningful problems. I’m particularly interested in:
- AI-powered knowledge management platforms
- Intelligent search and recommendation systems
- RAG and knowledge-enhanced AI applications
- Agent-based AI solutions for complex knowledge work
- Research and development in emerging AI technologies
Let’s Connect#
I believe in the power of knowledge sharing and collaboration. Whether you’re working on similar challenges, looking to explore new AI applications, or just want to discuss the latest developments in ML and knowledge graphs, I’d love to connect.
Feel free to reach out through any of my channels, or explore my projects on GitHub and posts on my website to see my work in action.
“The best solutions emerge when technical expertise meets deep understanding of real-world problems.”
