Tillmann Eitelberg
oh22information services GmbH
Königswinter, Germany
Actions
Tillmann Eitelberg is CEO and co-founder of oh22information services GmbH, which specializes in data management and data governance and offers its own cloud born data quality solution, HEDDA.IO.
Tillmann is a regular speaker at international conferences and an active blogger and podcaster at DECOMPOSE.IO. He has open sourced several SSIS components and is Co-Author of Power BI for Dummies (German Edition). Since 2013 Tillmann is awarded as Microsoft Data Platform MVP. He is a user group leader for the datamonster.io user group Rheinland (Cologne) and was a member of the Microsoft Azure Data Community Advisory Board.
Links
Badges
Area of Expertise
Topics
Garbage In, Hallucinations Out: Data Quality in the Age of AI
AI systems don’t fail silently, they fail creatively.
When data quality breaks down, models don’t just become inaccurate, they become unpredictable, biased, and operationally dangerous.
This session examines the fundamental effects of poor data quality on modern AI systems. It discusses how issues such as pipeline instability and missing context influence learning behavior during training, and why successful execution does not imply meaningful learning. Special attention is given to how data quality problems systematically amplify hallucinations, bias, and false confidence, particularly when AI systems are applied in business-critical contexts.
In the second part, the perspective shifts from AI as a victim of bad data to AI as a tool for improving data quality. The session shows how AI can be embedded into data engineering and data quality workflows to support profiling, semantic validation, and anomaly detection. We discuss how this changes the way pipelines, governance, and monitoring are designed when AI becomes part of the data engineering loop.
This talk focuses on control and trust in AI-driven systems. It helps practitioners understand where and why AI fails when data quality is insufficient, and how AI can be applied responsibly to improve data quality rather than undermine it.
Wrangling Data with Python – The Fabric Way
This one-day preconference offers a structured and hands-on introduction to working with data in Microsoft Fabric using Python. The focus is on understanding core components, applying Python in a Fabric-native context, and building reproducible, production-ready data workflows.
We begin with the foundational architecture of Microsoft Fabric: what a Lakehouse is, how Delta Tables are used, and how Spark operates within Fabric. Core Spark concepts will be explained without diving too deeply into internal mechanics, providing a clear understanding of how Fabric executes workloads.
Next, we shift to Python. After a brief recap of essential syntax, we focus on key programming principles relevant to data engineering. We also introduce essential libraries commonly used in Fabric-based workflows.
With this foundation in place, participants will move into guided hands-on exercises. These include developing data transformation and integration processes using both Python and Spark notebooks in Fabric. We’ll compare the notebook types, discuss the characteristics of different DataFrame implementations, and explain how Compute Units (CUs) influence execution performance.
In the next phase, we explore the benefits of combining Spark with other Fabric services, particularly Power BI. This includes an introduction to Semantic Link - demonstrating how transformed data can be connected to semantic models for reporting. Both technical integration and practical design considerations will be discussed.
The final part of the day focuses on development workflows using Visual Studio Code - locally and in the browser. We compare VS Code-based development with Fabric notebooks, highlight useful extensions, and explain scenarios where working outside the Fabric UI offers additional flexibility.
Data Observability with HEDDA.IO
Are you able to fully understand the state of your data in your systems? Are you able to evaluate the quality of your data in all processes?
In this session you will learn how easy it is to integrate HEDDA.IO into your existing processes, how it can provide you with a continuous view of the quality of your data and how it can inform you directly in the event of errors, warnings or major changes to individual parameters.
HEDDA.IO brings data observability to your process and platforms.
Data Quality Roundtrip 2023 in the Microsoft Data Platform
Data Quality Tools and Services in the Data Platform has always played a niche role. Anyone remembers DQS in SQL Server? As Data Governance plays more and more an important part in the strategy of Microsoft, we will run a demo packed session to show you what is currently available in Fabric, in Azure Services and as well in Microsoft Purview. We will discuss the pitfalls, the potential and have look to the benefits and possible strategy for your data estate
Integrate Data Quality into your processes
HEDDA.IO is a central data quality management solution that connects departments, data stewards and data engineers. It helps to easily integrate standardization, cleansing, matching and enrichment tasks into existing processes.
We use high-performance runners, in order to integrate HEDDA.IO into existing processes of a Modern Data Stack in the best possible way. The HEDDA.IO runners exist for pyspark and .NET and can therefore be used without any problems in Databricks, Azure Synapse Analytics and also in Visual Code together with Polyglot Notebooks.
Developers can thus extend their processes developed in Notebooks with HEDDA.IO. The full integration provides developers with IntelliSense and a meaningful widget for the results of an execution. All HEDDA.IO runners are designed to run on their respective systems (Databricks, Synapse, etc.) and not on the HEDDA.IO services.
Join us for this session to learn how you can quickly and easily add a high performance data quality solution to your existing and new processes.
Simplify Your Data Quality Processes
Identifying relevant data quality issues and creating awareness within organizations of the threats and costs is an important step in an enterprise-wide data quality strategy. This includes implementing systems and processes that enable organizations to continuously monitor the value of data, as well as creating processes to directly address errors in data processing.
With HEDDA.IO we focus on these challenges and provide a system to integrate data quality into almost all processes within your organization.
In this session we will show how quickly and easily Azure Synapse Analytics, Azure Databricks or .NET Interactive data processes can be extended with a Data Quality component and how you can add data observabilities to different scenarios.
Tillmann Eitelberg
oh22information services GmbH
Königswinter, Germany
Links
Actions
Please note that Sessionize is not responsible for the accuracy or validity of the data provided by speakers. If you suspect this profile to be fake or spam, please let us know.
Jump to top