Mischa Lisovyi Email and Phone Number

* Definition of tasks for the team in the agile work environment* Review of the team members' code

* An ETL pipeline with multi-process parallelisation using pandas + joblib* Store revenue modelling using LightGBM and Lasso regressions in python* User-facing dashboard using dash running in k8s using a docker image* A/B testing of the project success* Testing using pytest with 60% coverage over 25k+ LoC + code linting and type checks* Implementation of CI/CD pipelines in gitlab* Project templates for python libraries and apps using cookiecutter* Documentation using sphinx / mkdocs* Experiment tracking using MLFlow server in k8s* ETL scheduling and orchestration using Airflow

Big data analysis: - classification of hadronically-decaying heavy particles (Heidelberg University is one of the key players in the field) - identification of particular substructure/characteristic features of objects --- advanced DNN and gradient-boosted decision trees as well as simple decision-tree-like solutions; --- a framework to extract efficiency of the classifiers from data using template fit of simulations to the data;- calibration of jet energy scale using direct balance technique in Z+jet events:--- pre-selection of events on huge datasets of ~100 PB using high-throughput computing;--- local post-processing and data visualisation;Development of the data quality monitoring software for the new FPGA + custom CAM-based chips (Fast TracK trigger in the ATLAS experiment).Measurements of W & Z production cross sections @ LHCSupervision of ~100 bachelor and master students in particle-physics tutorials and lab courses. Supervision of PhD students in physics studies and detector performance work.

Involved in two large international collaborations: ZEUS experiment at the HERA collider and the ATLAS experiment at the LHC collider. Physics and performance studies:- electroweak bosons W,Z,gamma at LHC; - development, optimisation, maintenance and production of high-precision CPU-intensive theoretical predictions. Massively parallel computations with ~100.000 CPU hours. Significant improvements in the speed and precision of calculations allowed me to contribute to several published and ongoing analyses in ATLAS.;- implementation of silicon-strip geometry simulation in GEANT4 for detector upgrade;- maintained and developed the production system for the prompt calibration of the silicon strip ATLAS detector. Achieved a reduction in computation time by a factor more than 2.- coordination duties in the ZEUS physics group, supervising and organising the physics analyses in the collaboration up to the final steps.These activities required knowledge and application of sophisticated algorithms in data analysis as well as solid skills in programming (C/C++, python, fortran, shell), Monte Carlo simulations and statistics.

Working towards my PhD thesis in the international ZEUS experiment at the electron-proton collider HERA:- analysing data;.- responsible for Monte Carlo production with the high throughput computing (HTC) on the GRID;- coordinator of the overall Monte Carlo production system for event simulations.These activities required knowledge algorithms in data analysis as well as solid skills in programming (C/C++, shell), Monte Carlo simulations, large-scale parallel production system and statistics.