Florian Jetter
Software Engineer at QuantCo
Karlsruhe, Baden-Württemberg, Germany
Summary
🤩
Rockstar🎓
Top SchoolFlorian Jetter is a software engineer with 11 years of experience building reliable, distributed data systems and leading engineering teams from Karlsruhe, Germany. He blends hands-on backend and data engineering—contributing QA, bug fixes, and performance improvements to flagship open-source projects like pandas and Dask—with leadership roles at Coiled and QuantCo. Florian’s work shows deep expertise in distributed scheduling, Parquet/PyArrow handling, and elusive concurrency fixes (deadlocks, race conditions, work-stealing), reflecting a pragmatic approach to stability at scale. He started as a physicist (Master’s from Heidelberg) and brings that analytical rigor to software design and test automation. Notably, his contributions include optimizing categorical hashing and improving groupby/merge correctness, highlighting a focus on correctness as well as performance.
11 years of coding experience
9 years of employment as a software developer
Bachelor's degree Physics, Bachelor's degree Physics at University of KonstanzMaster's degree Physics, Master's degree Physics at Heidelberg University
Github Skills (28)
asynchronous10
debugging10
scheduled-tasks10
debug10
pytest10
pyarrow10
python10
async10
data-engineering10
testing10
pandas10
distributed-systems10
scheduler10
distributed-computing10
scheduling10
Programming languages (9)
Github contributions (5)
dask/distributed
Oct 2020 - Jan 2023
A distributed task scheduler for Dask
Role in this project:
Backend Developer
Backend DeveloperContributions:1940 reviews, 216 commits, 965 PRs in 2 years 3 months
Contributions summary:Florian primarily focused on enhancing the core functionality and stability of the Dask distributed task scheduler. Their contributions included fixing a regression in task stealing, removing hard-coded connection handshake timeouts, and addressing a deadlock in the work-stealing algorithm. The user also made improvements to the management of memory usage within the worker, fixed a potential deadlock related to task deserialization, and provided fixes for various race conditions in the system.
pythonschedulerdaskdistributed-computingdistributed-systems
dask/dask
Jan 2023 - Jan 2023
Parallel computing with task scheduling
Role in this project:Back-end Developer & Data Engineer
Back-end Developer & Data EngineerContributions:368 reviews, 4 commits, 240 PRs in 8 days
Contributions summary:Florian primarily contributed to bug fixes and feature implementations within the dask/dask repository. Their work focused on improving the handling of Parquet files within the Dask ecosystem, including addressing issues related to filesystem objects, improving metadata handling, and optimizing performance with the PyArrow library. Additionally, the user worked on optimizations for estimating python collection sizes and addressing bugs related to the groupby operation. Their contributions were across multiple modules and demonstrate a strong understanding of data processing and distributed computing.
pythonschedulingparallelnumpydask
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.