Software Engineering
Whether for web or desktop, I prefer a rigorous approach to software design utilizing TDD and design patterns
Hi there, I am Zuri
Lead engineer at ti&m
Leading, coding, and architecting software projects at ti&m, ETH Zurich alumnus, former data mining and HPC researcher, proud Father of Two and Specialist Officer for Computer Science in the Swiss Army
+7 years Experience
I've been working as a software engineer professionally since 2017, promoted to Senior in 2021, and then to Lead Engineer in 2024.
ETH Zurich Alumnus
I hold a Bachelor's degree in Computer Science from ETH Zurich
Embracing Contributions
I actively maintain several open-source projects, co-author research papers, and have begun to write blog posts in my free time.
What I do professionally
Software Architecture
Crafting software architecture, requires being up-to-date with old proven and latest trends alike
Data Engineering
At the moment I am engaged with data engineering projects in all sorts of scales in the cloud and on premise
Graph algorithms
Working on graph mining and other graph related problems has been one of my passions since my research days at ETH Zurich
High-performance computing
Engaging with HPC topics, preferably for shared memory architectures, has redefined my approach to coding
Blockchain
My past and ongoing participation in blockchain and DLT projects make appetite for more
My Tech Stack
Throughout my career, I've predominantly served as a full-stack engineer, employing C# (ASP.NET) and Angular alongside HTML, (S)CSS and TypeScript. I have deployed and maintained applications across diverse platforms such as Azure, AWS, Google Cloud, as well as different Kubernetes environments and specialized on-premises setups. Lately, I've been increasingly involved with the field of data engineering, fully engaged with tools and languages including Python, Polars, Spark, Trino, Iceberg, Superset, Kafka, various Databases and many others. In the past, I had also utilized Java and C++, including OpenMP and MPI. For a full tech breakdown, refer to my CV.
My Opensource Projects and Research Papers
Signalstory
Angular state management
Angular library that offers a fresh approach to state management while leveraging Angular signals. It provides various functionalities, ranging from fundamental repository services to advanced command orchestration and side-effect handling, streamlining the process while maintaining robustness. With features like native IndexedDB support, transactional undo/redo and many more: It's a lightweight and scalable solution for developers at any level.
GMS
Graph mine suite
GMS provides a comprehensive framework for the development and benchmarking of high-performance graph mining algorithms. Drawing upon extensive research, GMS defines representative problems, algorithms, and datasets. It offers an intuitive software platform for testing various elements of graph mining algorithms, including graph and set representations and algorithm subroutines.
PAPIW
PAPI wrapper
PAPIW simplifies "Performance Application Programming Interface" (PAPI) counter usage in C++ applications, offering a header-only solution for both sequential and parallel (OpenMP) software.
Parallel Graph Coloring
High-Performance Parallel Graph Coloring with Strong Guarantees on Work, Depth, and Quality
By introducing a novel relaxation of vertex degeneracy order, we develop the first parallel graph coloring heuristics with strong theoretical guarantees on work and depth and coloring quality. Our algorithms offer unmatched performance across various real-world graphs while ensuring polylogarithmic depth and superior coloring quality.
In Proceedings of the IEEE/ACM International Conference on High Performance Computing, Networking, Storage and Analysis (2020)
Read the paper GMS (Paper)
Enabling High-Performance and Programmable Graph Mining Algorithms with Set Algebra
We introduce a new solution to develop and benchmark graph mining algorithms. With over 40 baselines, GMS aids in improving efficiency, leveraging various building blocks for modularity. It enabled us to accelerate core graph mining tasks like degeneracy reordering (2x), maximal clique listing (9x), k-clique listing (1.1x), and subgraph isomorphism (up to 2.5x).
In Proceedings of the 47th International Conference on Very Large Data Bases (2021)
Read the paper SISA
Set-Centric Instruction Set Architecture for Graph Mining on Processing-in-Memory
Our paper offers a new approach to accelerate complex graph mining algorithms when using Processing-in-Memory (PIM) systems. By recognizing the significance of set operations and implementing a thoughtful cross-layer design, SISA introduces a novel programming paradigm and ISA extensions, surpassing traditional approaches by over 10x in certain scenarios.
In Proceedings of the 54th IEEE/ACM International Symposium on Microarchitecture (2021)
Read the paperContact me
Let's connect: Drop your info below and I'll reach out personally!