Saarland
Director II, leading the Streaming SQL and the core Flink Runtime team.
Building cloud-native Flink for Confluent Cloud.
Apache Flink is an open source system for expressive, declarative, fast, and efficient data analysis. Flink combines the scalability and programming flexibility of distributed MapReduce-like platforms with the efficiency, out-of-core execution, and query optimization capabilities found in parallel databases.
Optimizing Snowflake's distributed query engine for high concurrency and low latency workloads. Currently driving the development of Snowflake's Search Optimization Service.
Senior software engineer at data Artisans (now Ververica), where I helped creating the open-source stream processor Apache Flink. My main responsibility here was developing the state management parts (state storage, access, durability, consistency, rescaling, timer service, performance) for stateful stream-processing. Besides that, I also made significant contributions to the distributed coordination, scheduling, plugin mechanism and the streaming runtime in general.
Research focus: 1. Indexing in HDFS for Hadoop MapReduce: Invention and implementation of a distributed, low overhead indexing approach for HDFS. Technologies used include Java, HDFS, Hadoop MapReduce, Jenkins for CI, JVisualVM. 2. Index Structures for Main Memory Databases: Analysis, benchmarking and optimization of data structures for main memory databases. In particular, my work included B-trees, Tries (Judy Array, Adaptive Radix Tree), and Hash Tables with focus on optimization for modern hardware (cache efficient, SIMD, NUMA, OoOE). I also worked on efficient sorting/partitioning and joins. Technologies used include C/C++, ASM, Linux, perf, vTune, and IACA. Besides research, my work also included teaching activities such as helping students (B.Sc., M.Sc.) with their thesis work and preparing exercises for lectures. In particular, I was responsible for the practical project in each iteration of our database systems lecture (implementing a basic database in Java, ~80 students).