Gijs Koot

Gijs Koot

Lead data Scientist at Octo

Utrecht City, Utrecht Province, Netherlands, US




I am interested in data driven analysis, with a broad interest in applications, such as quantified self, media analysis, HR analytics, and smart cities. Ultimately, I want to understand and support people or organisations with statistical insights. To this end, I am always learning new theory and experimenting with new techniques.

In a job I am pragmatic, I like keeping things simple and I am good at choosing the right technology for the job at hand. In general I am analytical and will keep the bigger picture in mind.

My main interests are applied statistics, applied probability and machine learning algorithms.
In my spare time I like playing music and climbing.

Work Experience

Work Experience

  • Lead data ScientistOcto

    Jan, 2020 - Present

    Octo is building a platform for automated building managements driven by computer vision. In this team I am working on training models, geospatial calculations and growing the team.

    Technologies: Pytorch, AWS, PostgreSQL, Docker

  • Data Scientist, Urban Science - Axians

    2018-06-31 - Jan, 2020

    Urban Science is a unit within Axians, the software company. This new team applies data science to challenges that arise particularly in urban areas. The three focus markets are Energy, Mobility and Public Services.

    In this growing team I am working on

    • Developing and presenting new datadriven solutions in Urban Science
    • Applying statistical models and optimization algorithms
    • Projects for our customers, building cloud based datadriven applications
    • Help the team grow, in knowledge, skills and organisation

    Technologies: Python, PyMC3, rstan, Azure, scikit-learn, tensorflow, AWS, PostgreSQL, Docker

  • Data ScientistXomnia B.V.

    Jul, 2016 - Jun, 20181 year 11 months

    Working as a consultant and as a trainer. Specializing in text-mining and applied probabilty.

    The job involves

    • Data analysis, answering research questions, developing and evaluating algorithms
    • Presenting results in reports or talks
    • Preparing and giving trainings or lectures to other data scientists on topics such as textmining, timeseries
    • Interviewing and mentoring prospective data scientists
    • Creating and refining project proposals, specifying the proper approach and planning for a project

    Technologies; Python, PyMC3, Jupyter, RStudio, rstan, Linux, Numpy, Pandas, Scipy, Scikit, gensim.

  • Scientist / InnovatorTNO

    Oct, 2012 - Jun, 20163 years 8 months

    A part of the Data Science department, the Media Mining group applies techniques from Natural Language Processing, data science and computer vision to varying domains such as cybercrime, social media, smart cities, quantified self and media personalization. My work consists of researching, creating ideas and prototyping applications. I have worked in both research projects and commercial projects, as a scientist and a developer. Project teams are generally small and flat, and I was also involved in consultancy and strategy and as a project lead.

    Technologies: Python, R, AngularJS, d3.js, MongoDB, Elasticsearch, Linux, Git, Docker

  • Data analyst - OperationsThe Brighthouse

    Aug, 2011 - Jul, 201211 months

    The Brighthouse is a young company that gathers informations from surveys. I worked on building reports and architecture.

    Technologies: PostgreSQL, HTML, JavaScript

  • Mathematics TeacherVisser ´t Hooft Lyceum

    Jan, 2010 - Aug, 20111 year 7 months

    Teaching mathematics at various grades, students ranging from 12 to 17 years old. Autonomous teaching, course preparation, exam preparation. Course were mostly VWO level with 15-30 students. Also taught in English in bilingual groups.



  • Statistics
  • Machine Learning
  • Python
  • Pattern Recognition


  • Stochastics and Financial Mathematics, MSc., Universiteit van Amsterdam / University of Amsterdam

    Dec, 2003 - Dec, 2011

  • , Statistical Learning, Stanford Online - Lagunita

    Dec, 2016 - Dec, 2016

    With Distinction
  • , Probabilistic Graphical Models, Machine Learning, Coursera

    Dec, 2014 - Dec, 2016



  • Encouragement price on Accountability Hackathon 2016 Open State Foundation

    Awarded on:

    A jury awarded our prototype 'Comparea', which makes it possible to quickly make fair comparisons among Dutch municipalities, with a encouragement price. See https://accountabilityhack.nl/.



  • Authorship Analysis on Dark Marketplace Forums, Proceedings of the IEEE European Intelligence & Security Informatics Conference (EISIC), Manchester, UK.

    Published on: Jan 01, 2015

  • Foraging Online Social Networks, Intelligence and Security Informatics Conference (JISIC), 2014 IEEE Joint

    Published on: Sep 24, 2014

    A concise and practical introduction is given on Online Social Networks (OSN) and their application in law enforcement, including a brief survey of related work. Subsequently, a tool is introduced that can be used to search OSN in order to generate user profiles. Both its architecture and processing pipeline are described. This tool is meant as a flexible framework that supports manual foraging (and not replaces it). As such, we aim to bridge science's state-of-the-art and current security officer's practice. This article ends with a brief discussion on privacy and ethical issues and future work.

  • Privacy and User Trust in Context-Aware Systems,

    Published on: Jun 22, 2014

    Context-aware systems (CAS) that collect personal information are a
    general trend. This leads to several privacy considerations, which we outline in
    this paper. We present as use-case the SWELL system, which collects infor-
    mation from various contextual sensors to provide support for well-being at
    work. We address privacy from two perspectives: 1) the development point of
    view, in which we describe how to apply ‘privacy by design’, and 2) a user
    study, in which we found that providing detailed information on data collection
    and privacy by design had a positive effect on trust in our CAS. We also found
    that the attitude towards using our CAS was related to personal motivation, and
    not related to perceived privacy and trust in our system. This may stress the im-
    portance of implementing privacy by design to protect the privacy of the user.

  • Needle Custom Search, Advances in Information Retrieval, Springer Lecture Notes on Computer Science

    Published on: May 08, 2014

    Web search engines are optimized for early precision, which makes it difficult to perform recall-oriented tasks using these search engines. In this article, we present our tool Needle Custom Search. This tool exploits semantic annotations of Web search results and, thereby, increase the efficiency of recall-oriented search tasks. Semantic annotations, such as temporal annotations, named entities, and part-of-speech tags are used to rerank and cluster search result sets.