Artem Budishchev, Developer in Hamburg, Germany
Artem is available for hire
Hire Artem

Artem Budishchev

Verified Expert  in Engineering

Data Scientist and Developer

Hamburg, Germany
Toptal Member Since
May 18, 2020

Artem is a data scientist with a physics background and experience in machine learning, statistics, and programming in the fintech area. Artem is passionate about predictive and explanatory statistical modeling, especially unconventional methods like deep reinforcement learning, semi-supervised learning, among others.



Preferred Environment

Amazon Web Services (AWS), Sanic Web Server, Flask, Kubernetes, Docker, Azure, Seaborn, Matplotlib, Scikit-learn, TensorFlow, CatBoost, XGBoost, SQL, R, Python, Linux

The most amazing...

...thing I've created is a deep reinforcement learning agent for a fintech company, that brought a significant uplift in the desired KPI.

Work Experience

Principal Data Scientist

2019 - 2020
MASH Luxembourg
  • Built a fraud prevention model. A type of supervised learning model that was trained to directly optimize a KPI, which resulted in a significant monetary uplift.
  • Researched new methods for credit risk modeling, fraud detection, and reject inference.
  • Architected the data science infrastructure and workflow.
  • Created SLAs with QA, DevOps, and software development teams to streamline the model deployment process.
Technologies: SQL, Databricks, Azure, CatBoost, Python

Senior Data Scientist

2018 - 2019
collectAI GmbH
  • Delivered end-to-end data science projects, including core business ML solutions.
  • Researched and implemented state-of-the-art methods for the next generation of AI solution at the company.
  • Developed a message classification algorithm that allowed to significantly reduce operational workload for the clients.
  • Supported the sales team with ad hoc analyses and/or presentations for clients for customer retention and acquisition.
Technologies: Amazon Web Services (AWS), XGBoost, Kubernetes, Keras, TensorFlow, Python

Data Scientist

2016 - 2018
Kreditech Holding SSL GmbH
  • Built credit risk scoring models using gradient boosting.
  • Developed a programmatic solution for an optimal bidding strategy in search engine advertisement campaigns.
  • Designed and implemented the data processing pipeline for credit risk models.
  • Developed a type of monitoring software to identify problems in the conversion funnel using various KPIs, such as conversions, costs, CPA, and so on, that automatically notifies the stakeholders of any abnormalities in the KPIs.
Technologies: Python, Docker, XGBoost, R

Deep Reinforcement Learning (DRL) Agent for Customer Journey Optimization

A DRL model trained to optimize a client's KPIs.

I led the project completely end-to-end from its inception—researching and prototyping, deploying to production, and monitoring the performance. I supervised the software development team, which was supporting the project from a technical point of view. The model itself was implemented using Python and TensorFlow and deployed using Kubernetes on AWS.

Deep Contextual Bandit for Customer Journey Optimization

A deep contextual bandit trained to optimize a client-provided KPI.

I led the project end-to-end from ideation, researching and prototyping, deployment to production, and monitoring the performance. A customer's state feature representation was transformed by an autoencoder and subsequently clustered. The clusters were used as contexts in the bandit. The model was implemented using Python and TensorFlow and deployed to Kubernetes on AWS.

Fraud Detection Model

A fraud detection model that was developed for the Spanish market.

I collected, parsed, and cleaned the data from various sources (databases, flat files, cloud storage, and so on). We built a CatBoost-supervised learning model, which showed a healthy performance in the backtest. The model was trained to directly optimize a KPI, which led to a significant increase in monetary gain.

Text Message Classification Using Transfer Learning with BERT

I created a message (emails and SMS) classification model using a fine-tuned BERT model from Google, which significantly outperformed the baseline model (logistic regression and bag of words). Feel free to reach out for more details.

Attribution Modeling

I developed a sophisticated attribution modeling system that consists of four different approaches based on Shapley value, logistic regression, Markov chain model, and XGBoost model. The final attribution was calculated using the ensemble's prediction median.

Analytics for Cross-selling and Upselling Customers Across Various Products for a Fintech

I performed analytics in order to cross-sell and upsell customers across various lending products (POS, installment loans, and so on).

The project produced lists of customers likely to purchase the offered product or upgrade the existing one.


Python, R, SQL


Pandas, NumPy, TensorFlow, XGBoost, CatBoost, Scikit-learn, Matplotlib, Keras


Data Science


Linux, Amazon Web Services (AWS), Google Cloud Platform (GCP), Azure, Docker, Kubernetes, Databricks


Machine Learning, Artificial Intelligence (AI), Exploratory Data Analysis, Credit Risk, Deep Learning, Fintech, Data Analytics, Natural Language Processing (NLP), Attribution Modeling, Neural Networks, GPT, Generative Pre-trained Transformers (GPT), Deep Reinforcement Learning, Data Engineering, Sanic Web Server


BigQuery, Seaborn

Industry Expertise

Project Management, Marketing


PostgreSQL, Exasol, Google Cloud, Microsoft SQL Server



2011 - 2015

Ph.D. in Climate Change

VU University Amsterdam - Amsterdam, the Netherlands

2002 - 2007

Master's Degree in Radiophysics and Electronics

North-Eastern Federal University - Yakutsk, Russia

2006 - 2006

Diploma in Geophysics

Univesity Centre in Svalbard (UNIS) - Longyearbyen, Svalbard, Norway


Data Science Specialization