Machine Learning - Data Engineering - Data Science - MLOps

8/2022 – 11/2022

Data Engine to automate the analysis of the market key aspects around some trending
topics.
In GCP, a scheduled job runs twitter parametrized queries and saves them into BigQuery.
The data is processed with NLP techniques and the key topics are selected algorithmically.
The topics are then plotted in a word cloud which is accessed by the users over an API
deployed in GCP. The engine is deployed in a docker container and scheduled to run daily.
Designed and developed a Python-based data pipeline to automate the analysis of
trending technological topics, leveraging NLP for text processing and BigQuery for data
storage.
Containerized and deployed the solution in GCP, enabling daily automated
summarizations and delivering visual insights via an API-accessible word cloud.
Collaborated with the IT Innovation team lead to provide actionable insights by overcoming
challenges in processing unstructured Twitter data for accurate analysis.

Natural Language Processing, Text-Extraction, API-Entwickler, Docker

9/2020 – 11/2023

Pyspark Prozesse
Automatisierung von custom Dashboards
Data Engines in Docker AWS
Datawarehouse Lösungen in AWS Athena
Entwicklung von Dashboards

Projekte:
Log Server Analysis Issue Detection @ Hotusa. 2023/07 – 2023/11
Developed a robust data pipeline to unify diverse request and error logs, enabling the efficient computation of metrics such as mean response time, error rates, and request volumes across multiple granular dimensions.
The team built a dashboard with these data that empowered channel agents to identify integration issues with associates early, significantly reducing resolution times. Leveraged PySpark, AWS Athena, and AWS Glue for scalable data processing and analytics.

- Direct Channel Dashboard Automation @ Hotusa. 2023/02 – 2023/04 (Barcelona)
Built and deployed a robust data pipeline using PySpark, Athena, and Docker on AWS to integrate booking data and web logs for comprehensive analysis.
Defined dashboard requirements and supervised the development of a dashboard to be sent to hundreds of customers delivering monthly KPIs tailored to customer needs.
Ensured the successful delivery of the project by overseeing technically and organizationally all stages.

- Rate Controller Phase 2 @ Hotusa. 2021/12 – 2022/03
Using Pyspark, significant calculation extension and datamart population. See below project Rate Controller Dashboard. (AWS Sagemaker, AWS Glue, AWS Athena)

- Rate Controller System Dashboard @ Hotusa . 2020/10 – 2021/03
Designed and developed a near real-time processing engine to monitor price ratings and optimize market rates by ingesting over 300 million records daily. Implemented business rule processing using PySpark on AWS Glue, with data storage and querying facilitated through AWS Athena (Hive). The solution delivered actionable insights to support dynamic pricing strategies at scale.

Amazon Web Services (AWS), Big Data, Business Intelligence (BI), Data Warehousing, Docker, Python