Jakub Semric
Member Of Technical Staff at Pure Storage
Prague, Prague, Czechia
Summary
👤
Senior🎓
Top SchoolJakub Semric is a software engineer with nine years' experience building cloud-native storage, ranking systems, and data-processing services, currently contributing to Pure Storage's cloud block storage platform. He blends backend systems engineering with practical ML experience—from optimizing SQLite-backed HTTP caches to augmenting image preprocessing for RetinaNet—demonstrating both performance-minded and data-driven approaches. At Lucid and Blindspot he worked on microservices, cloud infrastructure and large-scale batch aggregation pipelines, showing a knack for reliable, concurrent systems. His academic background includes a Distinction MSc in Data Science from the University of Glasgow and a strong IT foundation from Brno University of Technology. Notably, his open-source contributions improve real-world robustness (thread-local SQLite connections, bulk_commit optimizations) and ML data augmentation pipelines that boost model resilience. He is based in Prague and combines production-grade engineering with hands-on research-driven ML practices.
9 years of coding experience
3 years of employment as a software developer
Bachelor's degree, Information Technology, A, Bachelor's degree, Information Technology, A at Brno University of TechnologyMaster's degree, Data Science, Distinction, Master's degree, Data Science, Distinction at University of Glasgow School of Computing Science
Stackoverflow
301reputation
156kreached
4answers
0questions
Github Skills (32)
caching10
performance-monitor10
sqlite10
python10
multithreading10
machine-learning10
keras10
performance-analysis10
thread10
hyper-threading10
computer-vision10
image-augmentation10
threaded10
disk-cache10
object-cache10
Programming languages (1)
Github contributions (5)
fizyr/keras-retinanet
Jul 2019 - Jul 2019
Keras implementation of RetinaNet object detection.
Role in this project:
ML Engineer
ML EngineerContributions:7 commits, 5 PRs, 7 comments in 4 days
Contributions summary:Jakub implemented visual effect functions within the `keras-retinanet` project. These functions were integrated into the image preprocessing and data generation pipelines, as well as the training and debug scripts. The primary focus was on image augmentation techniques such as adjusting contrast, brightness, hue, and saturation, contributing to the model's robustness and performance. The changes included modifications to existing utility functions and integration into the generator and training scripts.
deep-learningretinanetcomputer-visionobject-detectiontensorflow
requests-cache/requests-cache
Apr 2021 - Apr 2021
Persistent HTTP cache for python requests
Role in this project:Back-end Developer
Back-end DeveloperContributions:5 commits, 1 PR, 2 comments in 1 day
Contributions summary:Jakub focused on optimizing the SQLite backend for the requests-cache library. Their primary contributions involve implementing thread-local SQLite connections, addressing potential locking issues, and adding a `bulk_commit` context manager for improved performance. They also removed unnecessary locking on read operations and added no-op tests. This work likely improved the concurrency and efficiency of database interactions within the caching system.
redispythontransparentperformancesqlite
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.