Nicolaj Schmit, Developer in Copenhagen, Denmark
Nicolaj is available for hire
Hire Nicolaj

Nicolaj Schmit

Verified Expert  in Engineering

Data Scientist and Developer

Location
Copenhagen, Denmark
Toptal Member Since
October 16, 2020

Nicolaj's passion for making sense of data led him to his career today. He creates solutions from data—building comprehensive proof-of-concept models to give the customer the relevant insights and take the models to production in the cloud. Education-wise, Nicolaj has a background in mathematics, quantitative finance, machine learning, and economics.

Portfolio

Tv2/Danmark
Azure, Databricks, MLflow, Adobe, Google Ads, Optuna, PySpark...
City of Copenhagen
Data Science, SQL, Azure SQL, Azure Resource Manager (ARM), Azure Data Lake...
Nordea
SQL, Multidimensional Expressions (MDX), Oracle SQL, Python, Jira, Confluence...

Experience

Availability

Part-time

Preferred Environment

Confluence, Jira, Azure DevOps, Git, Databricks, Azure, Windows, PyCharm, Jupyter Notebook, Visual Studio Code (VS Code)

The most amazing...

...project I've developed was a near real-time document sorting model in the cloud and which was integrated with a document source system and email client.

Work Experience

Data Scientist | Machine Learning Engineer

2021 - PRESENT
Tv2/Danmark
  • Built various segments of the user base—based on user behavior, subscription details, and so on—from the website and video-streaming site.
  • Developed big-data flows using Delta Lake, Parquet files, and PySpark.
  • Built advanced user segments using various machine learning methods with different supervised methods.
  • Built framework and processes for productionizing and controlling the lifecycle management of ML models.
Technologies: Azure, Databricks, MLflow, Adobe, Google Ads, Optuna, PySpark, Azure Cloud Services, Machine Learning, Python, Pandas, Data Analysis, Scikit-learn, Machine Learning Operations (MLOps)

Data Scientist

2018 - 2021
City of Copenhagen
  • Developed POCs for machine learning models on behalf of a number of data science projects. The goal was to make smarter use of the data available to the city of Copenhagen in Python using packages such as Scikit-learn, PyTorch, and Spark ML.
  • Architected and built Azure cloud solutions using tools such as Azure functions, logic apps, key vaults, storage accounts, docker, and so on.
  • Automated deployment pipelines using Azure DevOps and ARM templates.
  • Built dashboards and visualizations either for presentations for stakeholders or for interactive use for clients using tools such as Power BI and Databricks.
  • Set up ELT/ETL pipelines using tools such as Azure SQL, data factory, data lakes, and PySpark.
Technologies: Data Science, SQL, Azure SQL, Azure Resource Manager (ARM), Azure Data Lake, Azure Logic Apps, Azure Functions, Azure DevOps, Databricks, Azure, PySpark, Python, OCR, Azure Cloud Services, Spark ML, Machine Learning, Generative Pre-trained Transformers (GPT), Natural Language Processing (NLP), GPT, Pandas, Data Analysis, Scikit-learn, Regex, Text Classification, NumPy

Product Owner | Assistant Market Risk Manager

2016 - 2018
Nordea
  • Built infrastructure components that satisfied the fundamental review of the trading book requirements; these included developing prototyped models using Python and Excel VBA.
  • Monitored and assessed the quality of market risk models in relation to P&L attribution eligibility tests using tools such as Excel, VBA, and Oracle SQL.
  • Specified core risk models for value-at-risk, expected shortfall, and internal model capital charge-calculations using tools such as VBA and Python.
  • Brokedown, prioritized, and estimated the time of tasks in my role as a product owner.
  • Organized the work of other developers using tools such as Jira and Confluence.
Technologies: SQL, Multidimensional Expressions (MDX), Oracle SQL, Python, Jira, Confluence, Excel VBA, Risk Management, Pandas, Data Analysis, NumPy

Student Assistant

2014 - 2016
Danish Medical Doctors Pension Fund
  • Extended and maintained parts of the data warehouse solution used for risk management, i.e., such as the setup of SSIS jobs, building stored procedures, and extending the SQL data model.
  • Maintained and developed the performance reports of the investments of the pension fund. The reports were developed with Excel, VBA, and SQL stored procedures.
  • Assisted more senior colleagues in the risk management and investment teams with ac-hoc tasks.
Technologies: Bloomberg, SQL Server Integration Services (SSIS), Microsoft Excel, Excel VBA, SQL, Risk Management, Data Analysis

Algorithm for Learning to Play Simple Games

https://github.com/ngs90/RLBanana
I developed an algorithm that learned to play the Unity game, Banana, based on deep reinforcement learning. The algorithm is an implementation of the ideas presented in the papers "Human-level Control Through Deep Reinforcement Learning" (2015) and "Prioritized Experience Replay" (2016). The algorithms were developed in Python and PyTorch.

Mail-sorting System

I built a mail-sorting system that can take a scanned document as input and send the scanned document to the expected receiver via email in near-real time. The classifier was built with Python, Databricks, Tesseract, Scikit-learn and was productionized in the Azure cloud. The deployment pipelines were automated with ARM templates.

Web Scraping Algorithm

I built a web scraping algorithm using Selenium and Beautiful Soup for the website, Boliga.dk, which is a website containing information about sold properties and houses. The crawler collects various bits of information about each sold house.

Languages

Python, SQL, Excel VBA, Regex, Python 3, R

Libraries/APIs

Pandas, Scikit-learn, PySpark, Spark ML, NumPy, Matplotlib, PyTorch, TensorFlow, Beautiful Soup, Terragrunt

Paradigms

Data Science, Azure DevOps

Platforms

Databricks, Azure, Azure Functions, Jupyter Notebook, Docker, Windows, RStudio, Visual Studio Code (VS Code)

Other

Machine Learning, Risk Management, Delta Lake, Azure Resource Manager (ARM), Quantitative Finance, Predictive Analytics, Data Analysis, Text Classification, CI/CD Pipelines, Bicep, Azure Data Lake, Multidimensional Expressions (MDX), Deep Reinforcement Learning, Tesseract, ARM, Deep Learning, Natural Language Processing (NLP), LSTM Networks, Gated Recurrent Unit (GRU), MLflow, Google Ads, Optuna, OCR, GPT, Generative Pre-trained Transformers (GPT), Machine Learning Operations (MLOps)

Storage

Azure Cloud Services, Oracle SQL, Azure SQL, Azure Table Storage, SQL Server Integration Services (SSIS)

Frameworks

Hadoop, Selenium

Tools

Azure Logic Apps, Confluence, Azure App Service, PyCharm, Git, Jira, Trello, LaTeX, Maple, Microsoft Excel, Bloomberg, Adobe, Terraform

2019 - 2020

Courses in Machine Learning in Computer Science

University of Copenhagen - Copenhagen, Denmark

2014 - 2016

Master's Degree in Mathematics and Economics

University of Copenhagen - Copenhagen, Denmark

2011 - 2014

Bachelor's Degree in Mathematics and Economics

University of Copenhagen - Copenhagen, Denmark

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