Author Image

Hi, I am Till Keyling

Till Keyling

Principle Software Architect at Payback GmbH

I am a passionate Software & Data Architect who loves planning and building data-intensive applications.
Coming from a computational social-science background (PhD), I studied large-scale information systems using machine-learning approaches for content analysis, built large-scale (web)-scrapers and created an OSS-Tool (Facepager).

In my free time I explore the mountains with my paraglider.

Skills

Experiences

1
Payback GmbH

Feb 2020 - Present

Munich, DE

Payback is a multi-national, data-driven marketing & loyalty-program that delivers marketing & loyalty-services to over 31 Million customers (in DE only) and hundreds of partners (POS-Partners, Online-Shops etc.) worldwide.

Principle Software Architect

Feb 2023 - Present

Responsibilities:
  • Shaping the technological vision and architectural strategy for the company’s Data- & AI-platform and various backend-systems.
  • Helping the company to adopt a Data Product/Data Mesh-thinking in a major organisational as well as a technological shift.
  • Assessing & driving the utilities and capabilities of AI-augmented software development practises.
Team Lead Software Engineering Data Science

Feb 2020 - Jan 2023

Responsibilities:
  • Leading, supporting and scaling two SE-Teams which build data-intensive applications (SAS, Python, Oracle) central to Paybacks operating model and delivering value to millions of customers on a daily basis
  • Responsible for a large software migration-project, shaping technological as well as organisational decisions for the future marketing-platform (Python, Containerization and Cloud-Services)
  • Conceptualization and Modernization of a new Data-Landscape within Payback (involving topics & concepts such as Data-Mesh & Domain-Driven-Design, Data Quality and Observability)

Munich, DE

As a digital group, ProSiebenSat.1 combines leading Entertainment brands with a with a strong Dating & Video and Commerce & Ventures portfolio under one roof.

Senior Data Scientist

May 2016 - Jan 2020

Responsibilities:
  • Prototyping & development of a web-application for the R&D division (Django, Java Spring, Elasticsearch), Steering & Monitoring an agile team (up to 6 developers) as a product owner to build an enterprise-level application for > 200 stakeholders within the company along the whole lifecycle
  • Trend detection, prediction and topic modelling on TV-usage/time-series & unstructured data in Python and R (LDA, Word2Vec, HLM)
  • Risk-assessment modeling for a Game-Show (loss prediction & insurance consulting)
  • Real-time analytics on large-scale HBBTV tracking-data in order to predict future market-shares of TV-programs (Webtrekk, Spark Streaming, Tableau)
  • Design and implementation of ETL-processes, constructing a DWH and building APIs for multiple divisions (Pentaho Kettle, Apache Airflow, Postgres, Django REST-Framework, Hive/Spark)
  • Architecture & development of a group-wide Data Science Workbench based on existing Hadoop infrastructure (CDH, JupyterHub, Yarn)
  • Technology consulting for the executive level (f.e. applications of blockchain-technologies in the area of license-management)
2

3

Munich, DE

The IfKW is internationally recognized for its excellence in research in a broad variety of areas including media and politics; the transformation of mediascapes; journalism, public relations and organizational communication; media use, reception and effects; as well as conflict, risk and health communication.

Research Assistant

Oct 2010 - Apr 2016

Responsibilities:
  • Worked in a multi-national DFG-research project in the field of political communication in the online world, with special focus on developing methods for the large-scale data collection and data analysis in social media such as YouTube, Twitter and Facebook.
  • Developing a toolkit for the automatization of social media data-collection via API’s (Facepager)
  • Supervised several bachelor- and master theses
  • Teaching activities such as lecturing a master seminar with a focus on digital methods.

Munich, DE

The IfKW is internationally recognized for its excellence in research in a broad variety of areas including media and politics; the transformation of mediascapes; journalism, public relations and organizational communication; media use, reception and effects; as well as conflict, risk and health communication.

Graduate Assistant

Jun 2007 - Jul 2009

Responsibilities:
  • Realized the technical implementation of data collection in online bulletin-boards for a multi-national research project
  • Quality control and evaluation of (manual) content analysis in the realm of online-communication/social media
4

5
solute GmbH

Mai 2007 - August 2008

Munich, DE

The solute GmbH offered a leading price-comparison platform (billiger.de)

Working Student/Intern

Mai 2007 - August 2008

Responsibilities:
  • Conducted and evaluated user surveys; executive summaries of usage data
  • Analysis of ad-campaigns (Google DFP)

Education

B.A & M.A in Communication Science
GPA: 1.2 (Dean´s List) out of 5
Publications:
Extracurricular Activities:
  • Graduate assistant (IfKW, LMU) in a DfG-funded reasearch project
  • Realized the technical implementation of data collection in online bulletin-boards for a multi-national research project Quality control and evaluation of (manual) content analysis in the realm of online-communication/social media
  • Held multiple presentations on international conferences

Publications & Talks

Lesson learned implementing a large-scale data mesh transformation at PAYBACK
Till Keyling Alessandro Confetti

Sharing lessons learned from PAYBACK’s early Data Mesh journey: where to start (tech, data, teams), how to build ownership without overhead, preventing fragmentation, and migrating from legacy warehouses to data products.

Lessons learned implementing a large-scale Data Mesh transformation
Alessandro Confetti Till Keyling Alexandra Lovin

A practical discussion on PAYBACK’s Data Mesh journey, moving beyond theoretical concepts to real-world implementation challenges and successes in a large-scale transformation.

Kollektives Gatekeeping. Die Herstellung von Publizität in Social Media [Collective Gatekeeping]

By developing a theoretical model of collective gatekeeping and supported by the empirical study of political video clips on YouTube, the author describes the production of public attention in social media and shows fundamental differences to traditional mass media. He shows that publicity in social media is a continuous struggle for attention, which leads to its strong unequal distribution and contradicts the ideal of democratization of political communication in and through social media.

News sharing in social media: A review of current research on news sharing users, content and networks
Anna Kümpel Veronika Karnowski Till Keyling

This article provides a review of scientific, peer-reviewed articles that examine the relationship between news sharing and social media in the period from 2004 to 2014. A total of 461 articles were obtained following a literature search in two databases (Communication & Mass Media Complete [CMMC] and ACM), out of which 109 were deemed relevant based on the study’s inclusion criteria. In order to identify general tendencies and to uncover nuanced findings, news sharing research was analyzed both quantitatively and qualitatively. Three central areas of research—news sharing users, content, and networks—were identified and systematically reviewed. In the central concluding section, the results of the review are used to provide a critical diagnosis of current research and suggestions on how to move forward in news sharing research.

Exemplification 2.0? The Roles of Direct and Indirect Social Information in Conveying Health Messages through Social Network Sites

The Internet in general, and social network sites in particular, provide great potential for communicating health issues. Different forms of online social information, such as user comments and likes, might foster the effectiveness of these communication channels. However, it is unclear whether these types of information affect people’s perceptions and judgments of health issues. Combining exemplification theory with research into online social information, we conducted a 2 × 3 × 2 experiment to address this question. We presented our participants (N = 577) with a discussion of influenza vaccination on mock-up of a Facebook page and varied the direct social information (user comments) and additional indirect social information (comment likes and post likes) given, to assess the impact on individuals’ perceptions of public behavior and risk, and on their personal attitudes and behavioral intentions. Our results showed that the participants relied on direct but not indirect social information when forming judgments.

Categorization and evaluation of different types of automated content-anlysis methodologies (NLP) and their applications within the social sciences.

Accomplishments

iSAQB® CPSA-F
Certible December 2022

The CPSA Foundation Level (CPSA‑F®) conveys the ability in a team to design and document an appropriate software architecture, to assure and evaluate quality, and to know the necessary tools, all that for properly described requirements and systems up to a medium size.

Apache Hadoop Basics
Linux Hotel January 2018

Introduction to Apache Hadopp, the MapReduce Framework, Apache Spark and Apache Hive.

Scrum Master (PSM I)
Scrum.org January 2017

Introduction to the foundations of Scrum & Agile Practices. Certification from Scrum.org as a Scrum Master (PSM I)