Otto Dobretsberger
Sr. Software Engineer
Actions
Dr. Otto Dobretsberger is a Senior Software Developer at CODE Consulting (part of the CODE Group). He holds a Bachelor's degree in Biomedical Informatics from UMIT, an institution in Austria, as well as a Master's and Ph.D. in Computer Science from the University of Houston. His professional expertise centers on high-performance computational analysis of genetic and genomic data, with extensive experience in C and C++. He specializes in WPF, C#, and C++ development, while maintaining proficiency in multiple additional programming languages. His project portfolio spans biomedical image processing, genetic and genomic data analysis, analytical software solutions for the oil and gas industry, and the development of custom controls. Dr. Dobretsberger has authored scientific papers and magazine articles and is a regular speaker at industry conferences.
Since 2012, he has served as an Adjunct Professor at the University of Houston, where he has taught a range of programming courses, including C++, Python, JavaScript, and Java.
In addition to his technical and academic roles, Dr. Dobretsberger organizes the Global AI Houston chapter, leading a local community of AI professionals and enthusiasts focused on knowledge sharing and advancements in artificial intelligence.
Outside of his professional pursuits, he is an accomplished powerlifter, actively training and competing. He holds multiple USPA National Records and IPL World Records.
The Silent Breach: Guardrails and Governance for a World Where Everyone Has an LLM
As AI adoption accelerates, organizations face a new class of security threats, many of them invisible until it’s too late. ShadowAI tools bypass governance. MCP servers expose dangerous capabilities. Poorly designed prompts leak secrets. And unguarded LLM workflows can quietly spread sensitive data across systems, logs, and third‑party APIs.
This session explores the emerging security landscape around AI usage, focusing on practical guardrails that prevent accidental data exposure, unauthorized agent behavior, and uncontrolled capability access. We’ll break down real‑world failure modes, demonstrate how insecure context design leads to cascading leaks, and show how to build robust, layered defenses that keep AI systems safe even under unpredictable user input.
Attendees will leave with a clear blueprint for securing AI deployments, from prompt‑level controls to architectural guardrails that protect MCP servers, agent frameworks, and enterprise data flows.
The .NET Express Lane: Rapid Full-Stack Development with Blazor and Azure
Get ready to experience the future of full-stack development, entirely within the seamless universe of Microsoft .NET! This talk is your ticket to building powerful, modern web applications without the constant context-switching to JavaScript. We'll harness the revolutionary power of Blazor to create rich, interactive frontends using C# and Razor syntax. We'll then architect a scalable, serverless backend with Azure Functions APIs and secure our data in the cloud with Azure SQL. The magic culminates as we use the deeply integrated tools in Visual Studio to deploy our entire application—frontend and backend—to Azure with breathtaking simplicity. Leave behind the complexity of juggling multiple languages and frameworks; we're about to unlock a new level of productivity and elegance by going all-in on .NET.
Prompt Alchemy: The Context Recipe for Maximizing AI Quality & Minimizing Cost
AI systems don’t get better by throwing more tokens at them. They get better when you engineer the right context. This session dives deep into the practical strategies behind min‑maxing AI token consumption while increasing output quality. We’ll go beyond traditional prompt engineering and explore reusable context blocks, hierarchical prompting, structured memory, and domain‑specific scaffolding that consistently produce higher‑quality results with fewer tokens.
Attendees will learn how to build modular context architectures, design reusable “prompt primitives,” and apply compression techniques that preserve meaning while reducing cost. Whether you’re building agents, workflows, or high‑volume LLM applications, this talk gives you the tools to scale intelligently, without sacrificing quality.
How Do AI Systems Like ChatGPT Work? A Clear Explanation for Everyone!
This session offers an accessible exploration of the inner workings of AI systems like ChatGPT, demystifying how these advanced language models generate human-like responses. We will break down the process of how large language models (LLMs) construct answers token by token, illustrating the step-by-step mechanics of language generation. Attendees will learn about the key components that drive these systems, including the role of training data, neural network architectures, and the probabilistic nature of response generation. We will also discuss how external factors, such as input prompts and fine-tuning, influence the model’s outputs, and explore the dependencies that shape their performance. Designed for a general audience, this talk avoids technical jargon while providing a clear, engaging look at the fascinating technology behind AI language systems, offering insights into their capabilities and limitations.
AI-Proofing Academia
The rapid emergence of Large Language Models (LLMs) has fundamentally altered the landscape of higher education, often leaving faculty feeling caught between a need for integrity and a lack of technical expertise.
We will move beyond the unreliable "cat-and-mouse" game of AI detection software and instead focus on proactive instructional design. Attendees will learn how to "AI-proof" assignments by shifting toward authentic, process-based assessments that prioritize critical thinking over static output. Furthermore, we will explore how faculty can use AI as a personal productivity partner—automating administrative tasks and generating lesson materials—while simultaneously teaching students how to use these tools ethically and professionally. By the end of this session, participants will have a clear roadmap for fostering academic honesty without banning the future.
Fortifying Your AI: Mastering Offensive and Defensive Strategies Against Prompt Hacking and Misuse
We’ll dive into the shadowy world of prompt hacking and unauthorized AI misuse, where attackers exploit vulnerabilities in large language models (LLMs) and integrated AI systems to manipulate outputs or extract sensitive data. Through real-world examples and hands-on demonstrations, you’ll uncover the tactics used in these attacks, from crafting malicious prompts to bypassing safety mechanisms. We’ll then pivot to robust defense strategies, exploring techniques to secure your code, harden applications, and implement safeguards like input validation, context-aware filtering, and monitoring. This session equips developers, security professionals, and AI enthusiasts with the knowledge to protect their systems while understanding the evolving threat landscape of AI misuse.
How Do Large Language Models (LLMs) – Such as ChatGPT – Actually Work?
Do you wonder how ChatGPT and other large language models (LLM) work? We’ll walk through a comprehensive example to illustrate the intricate process behind how LLMs like ChatGPT generate answers. Through this session, attendees will gain valuable insights into the underlying mechanisms that enable LLMs to understand and respond to human-generated queries. This session will demystify the complex algorithms and advanced techniques used by LLMs, providing a deeper understanding of their capabilities.
Through a carefully curated example (no code demonstrated), you'll witness firsthand how LLMs analyze and interpret the context, meaning, and nuances of questions, enabling them to provide contextually relevant and accurate responses. This interactive exploration will shed light on the state-of-the-art technology driving LLMs, allowing you to appreciate the incredible depth of their knowledge and reasoning abilities.
Whether you are an AI enthusiast, a researcher, or simply curious about the capabilities of LLMs, this webinar promises to be an enlightening and interactive experience. Don't miss this chance to uncover the secrets behind how LLMs transform raw input into insightful and meaningful answers.
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