Niyaz Orazbayev

Data Scientist at Travelier (formerly Bookaway Group)
  • Claim this Profile
Contact Information
us****@****om
(386) 825-5501
Location
Nur-Sultan, Kazakhstan, KZ
Languages
  • English Full professional proficiency
  • Russian Native or bilingual proficiency
  • Kazakh Native or bilingual proficiency

Topline Score

Topline score feature will be out soon.

Bio

Generated by
Topline AI

You need to have a working account to view this content.
Join now
You need to have a working account to view this content.
Join now

Credentials

  • Convolutional Neural Networks
    DeepLearning.AI
    Jan, 2023
    - Nov, 2024
  • AWS Machine Learning Foundations
    Udacity
    Oct, 2022
    - Nov, 2024
  • Big Data with PySpark track
    DataCamp
    May, 2022
    - Nov, 2024
  • Building Recommendation Engines with PySpark
    DataCamp
    Apr, 2022
    - Nov, 2024
  • Blue Prism Certified Associate Developer
    Blue Prism
    Mar, 2022
    - Nov, 2024
  • Databases and SQL for Data Science with Python
    IBM
    Oct, 2021
    - Nov, 2024
  • Improving Deep Neural Networks: Hyperparameter Tuning, Regularization and Optimization
    DeepLearning.AI
  • Neural Networks and Deep Learning
    DeepLearning.AI
  • Structuring Machine Learning Problems
    DeepLearning.AI

Experience

  • Travelier (formerly Bookaway Group)
    • Israel
    • Technology, Information and Internet
    • 1 - 100 Employee
    • Data Scientist
      • Feb 2023 - Present

      Working in SG subsidiary TravelTech company Working in SG subsidiary TravelTech company

  • Scholabot
    • Singapore, Singapore
    • Machine Learning Engineer
      • Oct 2022 - Jan 2023

      • Built math question difficulty level classifier using BERT for Singaporean edtech startup • Built math question difficulty level classifier using BERT for Singaporean edtech startup

  • KPMG Caucasus and Central Asia
    • Kazakhstan
    • Financial Services
    • 400 - 500 Employee
    • Data Scientist
      • Nov 2021 - Oct 2022

      • Achieved F1 score 0.8 for employee churn prediction model (HR analytics) • Power BI team lead for the project on building BI system and DWH for the largest power grid operating company in Kazakhstan (Power BI Report Server, SAP BW). • Tender classification (NLP) which saved 5.5 hours per day of manual work • Parsed tenders from various government websites (selenium) • Achieved F1 score 0.8 for employee churn prediction model (HR analytics) • Power BI team lead for the project on building BI system and DWH for the largest power grid operating company in Kazakhstan (Power BI Report Server, SAP BW). • Tender classification (NLP) which saved 5.5 hours per day of manual work • Parsed tenders from various government websites (selenium)

  • Bank CenterCredit
    • Banking
    • 700 & Above Employee
    • Data Scientist
      • Mar 2020 - Oct 2021

      • Built ML based fraud detection model that helped to detect 8 massive fraudsters • Built web app - interactive network visualization of transactions (Networkx, Plotly, Dash) that allowed convenient investigation of fraudulent transactions • Found hundreds of previously unknown employee connections represented by graph (networkx) • Clustered bank clients using community detection algorithms • Provided various calculations on the probability of operational risk incidents occurrence • Prepared Power BI reports for the management board Show less

  • Sberbank
    • Banking
    • 700 & Above Employee
    • Data Science trainee
      • Feb 2020 - Mar 2020

      • Parsed data using python libraries (bs4) • Built ML models (regression, classification, clustering) • Finished final team project - 1st place on kaggle competition leaderboard • Parsed data using python libraries (bs4) • Built ML models (regression, classification, clustering) • Finished final team project - 1st place on kaggle competition leaderboard

Education

  • Kazakh-British Technical University
    Bachelor of Applied Science - BASc, Applied Mathematics
  • Udacity
    Nanodegree, AWS Machine Learning Engineer
    2022 - 2023

Community

You need to have a working account to view this content. Click here to join now