Optimization of the exisiting person identifcation products
Data Scientist/ Analyst
Data Scientist/ Analyst
Optimization of the exisiting person identifcation products using data from the Postgres backend and JSON frontend to reduce the volume of fraud in e-commerce purchases and false positives of real consumers
Development of a new credit scoring and fraud detection product focusing on the energy and utilities sector, utilizing regression and gradient boosted models as well as techniques to handle highly imbalanced datasets such as the imblearn library in Python
Benchmarking models such as XGBoost, AdaBoost with integrated undersampling and VotingClassifer, against the established Logistic Regression method using metrics such as the f1-score and Gini-score.
Usage of the libraries pandas for small scale datasets and polars, pySpark for larger than RAM datasets, with sklearn, Pydantic, FastAPI, Docker and Git for model building, code refactoring and deployment
Creditreform Boniversum
Neuss
6 months
2024-05 - 2024-10
Supporting role within a project aiming to optimize marketing expenditure
Consultant Data Science
Consultant Data Science
Supporting role within a project aiming to optimize marketing expenditure per year in terms of sales volume
Derivation of concrete proposals for action that increases sales volume with the same marketing budget
Presentation of a new marketing strategy which focuses marketing spend on customer engagement channels resulting in higher sales volumes and a increased utilization of the marketing budget
elanyo GmbH
Stuttgart
7 months
2022-10 - 2023-04
Performed variable analysis of omnichannel channels
Working Student Data Science
Working Student Data Science
Performed variable analysis of omnichannel channels using machine learning models such as XGBoost and linear regressions as well as data science techniques.
Utilized the SHAP library to extract the top sales and marketing channels indicative of a successful customer acquisition.
AstraZeneca GmbH
Hamburg
1 year 1 month
2021-10 - 2022-10
Analysis of horticultural and chemical data of different materials
Data Scientist
Data Scientist
Analysis of horticultural and chemical data of different materials using data science libraries in Python such as pandas, polars, numpy, seaborn and the SHAP package
Development of regression and classifcation models using regression models and non-linear approaches such as gradient boosting and random forests
Development of a holisitic dashboarding solution for palm oil mills to visualize critical processes and metric tracking to reduce downtimes and help meet client KPIs
Provided actionable recommendations of suitable materials as a peat moss substitue to end clients based on technical analytics and data science performed
Skanda Consultancy
Kuala Lumpur
Aus- und Weiterbildung
Aus- und Weiterbildung
4 months
2022-03 - 2022-06
Data Science Program
Spiced Academy, Berlin
Spiced Academy, Berlin
Theory and application of data science and machine learning in areas such as classifcation, regression, clustering, recommender systems and image classifcation.
Practical case studies of Docker, Git, MongoDB, web scraping and programming best practices.
3 years
2018-10 - 2021-09
Industrial Engineering & Management
Bsc, Jacobs University Bremen
Bsc
Jacobs University Bremen
Final grade - 1.88
Kompetenzen
Kompetenzen
Top-Skills
Machine LearningData Scientist
Produkte / Standards / Erfahrungen / Methoden
Git
Machine Learning
PostgreSQL
R
MS Offce
Python
MS PowerBI
Docker
Profile
Competent data scientist and analyst with experience in generating insights and extracting value from data with analytics and machine learning, driving strategic and business goals.
Einsatzorte
Einsatzorte
Deutschland, Schweiz, Österreich
möglich
Projekte
Projekte
1 year 6 months
2023-05 - now
Optimization of the exisiting person identifcation products
Data Scientist/ Analyst
Data Scientist/ Analyst
Optimization of the exisiting person identifcation products using data from the Postgres backend and JSON frontend to reduce the volume of fraud in e-commerce purchases and false positives of real consumers
Development of a new credit scoring and fraud detection product focusing on the energy and utilities sector, utilizing regression and gradient boosted models as well as techniques to handle highly imbalanced datasets such as the imblearn library in Python
Benchmarking models such as XGBoost, AdaBoost with integrated undersampling and VotingClassifer, against the established Logistic Regression method using metrics such as the f1-score and Gini-score.
Usage of the libraries pandas for small scale datasets and polars, pySpark for larger than RAM datasets, with sklearn, Pydantic, FastAPI, Docker and Git for model building, code refactoring and deployment
Creditreform Boniversum
Neuss
6 months
2024-05 - 2024-10
Supporting role within a project aiming to optimize marketing expenditure
Consultant Data Science
Consultant Data Science
Supporting role within a project aiming to optimize marketing expenditure per year in terms of sales volume
Derivation of concrete proposals for action that increases sales volume with the same marketing budget
Presentation of a new marketing strategy which focuses marketing spend on customer engagement channels resulting in higher sales volumes and a increased utilization of the marketing budget
elanyo GmbH
Stuttgart
7 months
2022-10 - 2023-04
Performed variable analysis of omnichannel channels
Working Student Data Science
Working Student Data Science
Performed variable analysis of omnichannel channels using machine learning models such as XGBoost and linear regressions as well as data science techniques.
Utilized the SHAP library to extract the top sales and marketing channels indicative of a successful customer acquisition.
AstraZeneca GmbH
Hamburg
1 year 1 month
2021-10 - 2022-10
Analysis of horticultural and chemical data of different materials
Data Scientist
Data Scientist
Analysis of horticultural and chemical data of different materials using data science libraries in Python such as pandas, polars, numpy, seaborn and the SHAP package
Development of regression and classifcation models using regression models and non-linear approaches such as gradient boosting and random forests
Development of a holisitic dashboarding solution for palm oil mills to visualize critical processes and metric tracking to reduce downtimes and help meet client KPIs
Provided actionable recommendations of suitable materials as a peat moss substitue to end clients based on technical analytics and data science performed
Skanda Consultancy
Kuala Lumpur
Aus- und Weiterbildung
Aus- und Weiterbildung
4 months
2022-03 - 2022-06
Data Science Program
Spiced Academy, Berlin
Spiced Academy, Berlin
Theory and application of data science and machine learning in areas such as classifcation, regression, clustering, recommender systems and image classifcation.
Practical case studies of Docker, Git, MongoDB, web scraping and programming best practices.
3 years
2018-10 - 2021-09
Industrial Engineering & Management
Bsc, Jacobs University Bremen
Bsc
Jacobs University Bremen
Final grade - 1.88
Kompetenzen
Kompetenzen
Top-Skills
Machine LearningData Scientist
Produkte / Standards / Erfahrungen / Methoden
Git
Machine Learning
PostgreSQL
R
MS Offce
Python
MS PowerBI
Docker
Profile
Competent data scientist and analyst with experience in generating insights and extracting value from data with analytics and machine learning, driving strategic and business goals.
Vertrauen Sie auf Randstad
Im Bereich Freelancing
Im Bereich Arbeitnehmerüberlassung / Personalvermittlung