Daniel Kostic
Verified Expert in Engineering
Machine Learning Developer
Berlin, Germany
Toptal member since September 7, 2021
Daniel is a data science and machine learning specialist with more than two years of academic and practical experience working as an end-to-end data scientist. In most of his previous projects, he worked on relational data; however, he would be happy to take on a challenge based on unstructured data, images, and video.
Portfolio
Experience
- SQL - 5 years
- Python - 5 years
- Scikit-learn - 4 years
- Pandas - 4 years
- Docker - 3 years
- Machine Learning - 3 years
- Data Science - 3 years
- XGBoost - 2 years
Availability
Preferred Environment
Python, PyCharm, Jupyter Notebook, Scikit-learn, XGBoost, NumPy, Pandas, SQL, Docker, Git
The most amazing...
...project I worked on was analyzing the citation network of all CS publications in the past 50 years and making an ML model to predict the success of scientists.
Work Experience
Data Analyst
Auto1 Group
- Built and maintained over 15 reports and dashboards used daily by key stakeholders to track business performance and aid informed decision-making.
- Led a workflow transition in my team to include version control, bug tracking, and automated testing. Structured team task management and documentation by introducing Jira to the workflow.
- Developed a web app using Django that handles uploading and processing of all refurbishment invoices. The tool is used by all European teams internally to handle invoices from external partners.
- Automated various business processes, emails, and employee activities with Python pipelines saving hundreds of hours of employee time daily.
- Implemented a rule-based prediction model that uses available logistics data to estimate the delivery ETA of cars to the customer. The prediction is communicated to the customer as well as used internally.
Data Scientist
GESIS Leibniz Institute for the Social Sciences
- Developed an ML model for predicting the success of scientists based on features extracted from bibliometric data. Performed initial exploratory analysis, feature creation and extraction as well as gender inequality analysis.
- Co-authored a paper analyzing gender inequalities and success predictability of authors in computer science.
- Built a web app using Django and Redwood, a framework for running experiments in social science. After implementation, I deployed the solution, ran the experiments, and analyzed the collected data.
Experience
Predicting Success in Computer Science Academia
I was in charge of data collection, feature creation and extraction, model training, and result visualization. I participated in writing the paper as a co-author.
Dashboard for Managing Orders to External Partners
Refurbishment Invoicing Web Tool
Education
Master's Degree in Web Science
University of Koblenz-Landau - Koblenz, Germany
Bachelor's Degree in Computer Science
University of Belgrade - Belgrade
Certifications
Deep Learning Specialization
Coursera
Skills
Libraries/APIs
Scikit-learn, Pandas, XGBoost, NumPy, NetworkX, Google Sheets API, Matplotlib
Tools
Git, PyCharm, Jupyter, LaTeX, Redash
Languages
Python, SQL, JavaScript
Platforms
Jupyter Notebook, Docker, Web, Amazon Web Services (AWS), Linux
Frameworks
Django
Storage
Redshift
Other
Data Science, Machine Learning, Statistics, Data Analysis, Statistical Analysis, Data Analytics, Statistical Modeling, IT Management, Bayesian Statistics, Deep Learning, Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs)
How to Work with Toptal
Toptal matches you directly with global industry experts from our network in hours—not weeks or months.
Share your needs
Choose your talent
Start your risk-free talent trial
Top talent is in high demand.
Start hiring