Profileimage by Sebastian Napiorkowski Big Data Solutions Architect & Machine Learning Engineer from London

Sebastian Napiorkowski

available
Contact

Last update: 02.01.2022

Big Data Solutions Architect & Machine Learning Engineer

Graduation: M. Sc. Computer Science, RWTH Aachen University
Hourly-/Daily rates: show
Variiert mit dem Ort und den Projektspezifikationen
Languages: German (Native or Bilingual) | English (Full Professional)

Skills

Backend: Python, Django, Java, Jersey, Jetty, Java Script ES6, Node.js, Express, Android, Cloudera/Hortonworks Stack, Apache NiFi, Apache Kafka, Apache Spark
Databases: MySQL, PostgreSQL Apache HBase, Apache Hadoop, MangoDB, CouchDB, ElasticSearch, ELK-Stack, Cassandra, Memcached, Redis
Frontend: HTML 5, React, Java Script ES6, jQuery, d3.js, Android, Kibana, Tableau, Amazon Quicksight
Deployment: Docker, Amazon AWS ecosystem, EC2, S3, Google App Engine, Heroku, Apache, Linux Server
Data Science: numpy, sci-kit-learn, sci-py, pandas, machine learning, feature engineering, natural language processing, clustering
Miscellaneous: Git, Svn, Jira, Confluence, Slack, Google Analytics, IoT

Project history

07/2019 - 09/2020
Big Data Solution Architect & Information Retrieval Specialist
DAX30 group (Ludwigshafen am Rhein) (>10.000 employees)
Automotive and vehicle construction
Conceptualization, architecture and development of a scalable Big-Data solution (as a R&D Datalake use case) for mass indexing file contents using bleeding edge natural language processing and machine learning algorithms on the Cloudera Hadoop Stack (HDP and HDF), Palantir Foundry and Kubernetes Deployment in Microsoft Azure.

Technologies used
  • NiFi and MiNiFi for ETL
  • Apache Spark Processing (Java, Scala & Python)
  • Apache HBase
  • Elasticsearch Stack
  • Django Backend
  • React Frontend
Features comprising i.a.
  • Raw text extraction from various file types
  • Language dependent indexing
  • Clustering Approaches (i.a. Latent Dirichlet Allocation, Latent Semantic Indexing, doc2vec)
  • Parse unstructured data into structured data
  • Named Entity Recognition (i.a. chemical entities) using Neural Networks
  • Entity Linking (Distant Knowledge)
  • Molecular Substructure Search

Contact form

Contact details