Jedrzej Kardach, Developer in Poznań, Poland
Jedrzej is available for hire
Hire Jedrzej

Jedrzej Kardach

Verified Expert  in Engineering

Data Scientist and Machine Learning Developer

Location
Poznań, Poland
Toptal Member Since
May 22, 2023

Jedrzej is an accomplished full-stack data scientist and ML engineer with five years of experience. With strong Python back-end engineering skills, he excels in fast-paced, customer-value-centric work environments. Jedrzej collaborated with Princeton University's ORFE department to develop cutting-edge simulation environments to build sponsored search auction algorithms. Additionally, he has successfully delivered multiple NLP-based classification algorithms and reinforcement learning solutions.

Portfolio

Kalepa
Algorithms, Data Mining, Decision Tree Regression, Data Structures...
Booksy
Python, Python 3, BigQuery, Google BigQuery, Pandas, NumPy, Matplotlib...
Ora Ai
Algorithms, Python 3, Python, NumPy, Pandas, Scikit-learn, TensorFlow...

Experience

Availability

Part-time

Preferred Environment

MacOS, Jupyter Notebook, PyCharm, Python, Django, Machine Learning, Deep Learning, TensorFlow, Scikit-learn, Pandas

The most amazing...

...thing I have accomplished is obtaining a 5-fold improvement in the performance of a model created by a specialist with two levels of seniority above me.

Work Experience

Machine Learning Engineer

2022 - 2023
Kalepa
  • Delivered an NLP binary classification model involving both topic classification and named entity recognition, achieved a 5-fold improvement in F1 score over the predecessor, defined target variables, and supervised labeling efforts.
  • Rolled out multiple multi-label NLP classification models for drawing insights about business entities from large unstructured textual data.
  • Built an information retrieval algorithm from large, unstructured textual data.
  • Maintained the back-end infrastructure for models and deployed them using serverless AWS Lambda, AWS Step Functions, and Amazon SageMaker.
  • Integrated ChatGPT API into the company's client-facing services, which included prompt engineering to maximize ChatGPT's utility for a given use case.
Technologies: Algorithms, Data Mining, Decision Tree Regression, Data Structures, Deep Learning, TensorFlow, Natural Language Toolkit (NLTK), Natural Language Processing (NLP), Python, Python 3, Amazon Web Services (AWS), AWS Lambda, AWS Step Functions, PostgreSQL, SQL, Machine Learning, Scikit-learn, NumPy, Pandas, Seaborn, Matplotlib, Hyperparameters, Random Forests, Support Vector Machines (SVM), Logistic Regression, Linear Regression, Web Scraping, Data Inference, ChatGPT, Amazon SageMaker, Git, GitHub, Jira, Confluence, Jupyter Notebook, ETL, Classification, Text Classification, Text Mining, APIs, Jupyter, Requests, Information Retrieval, Artificial Intelligence (AI), Data Science, Data Analysis, Data Analytics, Data Modeling, Data Visualization, Data Collection, OpenAI GPT-3 API, Language Models, Classification Algorithms, Generative Pre-trained Transformers (GPT), GPT, Generative Pre-trained Transformer 3 (GPT-3), Pytest, Keras, Dashboards, Analytical Dashboards, Programming, Integration, OpenAI GPT-4 API, TensorFlow Deep Learning Library (TFLearn), Modeling, Exploratory Data Analysis

Data Scientist

2020 - 2022
Booksy
  • Developed an event-based product monitoring architecture for parts of the application, designed performance monitoring for A/B tests, and suggested improvements to the product resulting in a 30% increase in B2C acquisitions.
  • Designed and implemented ETL pipelines and machine learning models for churn prediction and user clustering.
  • Delivered business intelligence dashboards using Microsoft Power BI, SQL, and BigQuery.
Technologies: Python, Python 3, BigQuery, Google BigQuery, Pandas, NumPy, Matplotlib, Scikit-learn, TensorFlow, ETL, Jira, Confluence, User Monitoring, Firebase, Microsoft Power BI, Business Intelligence (BI), Classification, Machine Learning, Google Cloud Platform (GCP), Support Vector Machines (SVM), Clustering, K-means Clustering, Logistic Regression, Linear Regression, Decision Trees, Random Forests, XGBoost, Artificial Intelligence (AI), Data Science, Data Scientist, Data Analysis, Data Analytics, Predictive Modeling, Predictive Analytics, Forecasting, Data Modeling, Data Visualization, Data Collection, Classification Algorithms, Keras, Dashboards, Analytical Dashboards, Interactive Dashboards, Programming, TensorFlow Deep Learning Library (TFLearn), Modeling, Exploratory Data Analysis, Consumer Behavior

Data Scientist

2018 - 2022
Ora Ai
  • Conducted a yearlong R&D process with the operations research and financial engineering department at Princeton University, which focused on developing a simulator of Google Ads environment for policy testing in a reinforcement learning setting.
  • Built countless machine learning-based modules to forecast time series, solve regression and classification problems, or approximate values of specific metrics. All these modules aid the main bidding algorithm in making final bidding recommendations.
  • Participated in the development of AI technology that optimally manages the bidding process in sponsored search auctions alongside researchers from Princeton University.
  • Collaborated in the development of a simulator for a hotel economic environment using the Monte Carlo method that allowed testing algorithms for automated marketing.
  • Co-developed reinforcement learning algorithms for automated marketing in hotels. This included setting an optimal price on special offers and choosing the right communication channel and time.
  • Co-created a web application that facilitates the supervision of digital marketing campaigns as well as the AI that manages these campaigns.
  • Developed a Django REST API and an ETL pipeline for real-time data collection from the hotel management software.
Technologies: Algorithms, Python 3, Python, NumPy, Pandas, Scikit-learn, TensorFlow, Ad Optimization, Optimization, Google Ads, Google AdWords, Real-time Bidding (RTB), Online Bidding, Simulations, Reinforcement Learning, Time Series, Linear Regression, Logistic Regression, Decision Trees, Random Forests, XGBoost, Support Vector Machines (SVM), Clustering, K-means Clustering, Django, Django ORM, Django REST Framework, REST, Postman, Docker, Docker Hub, Docker Cloud, Docker Compose, Amazon Web Services (AWS), Amazon EC2, Amazon S3 (AWS S3), PostgreSQL, MySQL, SQL, SQLite, SQLAlchemy, Dash, Web Dashboards, Dashboard Design, Plotly, Seaborn, Matplotlib, Flask, SciPy, Jira, Learning, Bayesian Statistics, Bayesian Inference & Modeling, Operations Research, Bitbucket, Git, Artificial Intelligence (AI), Data Science, Data Scientist, Data Analysis, Data Analytics, Predictive Modeling, Predictive Analytics, Forecasting, Data Modeling, Data Visualization, Data Collection, Classification Algorithms, Keras, Data-driven Marketing, Pricing Models, Dashboards, Analytical Dashboards, Interactive Dashboards, Programming, Modeling, Exploratory Data Analysis, Consumer Behavior

Optimal Motivation Scheme System for a Call Center

https://www.cambridge.org/engage/miir/article-details/61831286ad7f7c742d5411f5
Co-designed a system for optimal task allocation for employees of a large call center that resulted in the publication of a paper. The objective was to maximize employee productivity while keeping their stress at an acceptable level.

Languages

Python, Python 3, SQL, Java

Libraries/APIs

Scikit-learn, Pandas, NumPy, Matplotlib, XGBoost, TensorFlow, Django ORM, SQLAlchemy, Keras, TensorFlow Deep Learning Library (TFLearn), Natural Language Toolkit (NLTK), Google AdWords, SciPy, Requests

Tools

Seaborn, Jupyter, PyCharm, Git, GitHub, BigQuery, Postman, Bitbucket, AWS Step Functions, Amazon SageMaker, Jira, Confluence, Microsoft Power BI, Docker Hub, Docker Compose, Plotly, Pytest

Paradigms

Data Science, ETL, Business Intelligence (BI), REST, Linear Programming

Platforms

MacOS, Jupyter Notebook, Amazon Web Services (AWS), AWS Lambda, Firebase, Google Cloud Platform (GCP), Docker, Amazon EC2, NoCodeAPI

Other

Data Mining, Machine Learning, Natural Language Processing (NLP), Logistic Regression, Linear Regression, Data Inference, Classification, Text Classification, Text Mining, Simulations, Clustering, K-means Clustering, Learning, Artificial Intelligence (AI), Data Scientist, Data Analysis, Data Analytics, Predictive Modeling, Data Modeling, Classification Algorithms, Programming, Modeling, Algorithms, Random Forests, Support Vector Machines (SVM), Decision Trees, Gradient Boosted Trees, Time Series, Time Series Analysis, Data Structures, Deep Learning, Hyperparameters, Web Scraping, APIs, Google BigQuery, User Monitoring, Real-time Bidding (RTB), Online Bidding, Reinforcement Learning, Bayesian Statistics, Predictive Analytics, Forecasting, Data Visualization, OpenAI GPT-3 API, Language Models, GPT, Generative Pre-trained Transformer 3 (GPT-3), Data-driven Marketing, Pricing Models, Dashboards, Analytical Dashboards, OpenAI GPT-4 API, Exploratory Data Analysis, Linear Algebra, Statistics, Statistical Methods, Probability Theory, Optimization, Partial Differential Equations, Finance, Mathematics, Analysis, Mathematical Analysis, Cryptography, Discrete Mathematics, Operations Research, Sorting Algorithms, Random Forest Regression, Support Vector Regression, Decision Tree Regression, ChatGPT, Ad Optimization, Google Ads, Dash, Web Dashboards, Dashboard Design, Bayesian Inference & Modeling, Information Retrieval, Data Collection, Generative Pre-trained Transformers (GPT), Interactive Dashboards, Integration, Consumer Behavior

Frameworks

Django, Django REST Framework, Flask

Storage

PostgreSQL, MySQL, SQLite, Docker Cloud, Amazon S3 (AWS S3)

Industry Expertise

Project Management

2017 - 2018

Master's Degree in Mathematics

The London School of Economics and Political Science (LSE) - London, United Kingdom

2013 - 2017

Bachelor's Degree in Mathematics

King's College London - London, United Kingdom

AUGUST 2013 - PRESENT

Project Management Advanced Topics

Project Management Institute (PMI)

AUGUST 2013 - PRESENT

Project Management Fundamentals

Project Management Institute (PMI)

Collaboration That Works

How to Work with Toptal

Toptal matches you directly with global industry experts from our network in hours—not weeks or months.

1

Share your needs

Discuss your requirements and refine your scope in a call with a Toptal domain expert.
2

Choose your talent

Get a short list of expertly matched talent within 24 hours to review, interview, and choose from.
3

Start your risk-free talent trial

Work with your chosen talent on a trial basis for up to two weeks. Pay only if you decide to hire them.

Top talent is in high demand.

Start hiring