Aldan Creo
Technology Research Specialist @ Accenture Labs
Dublin, Ireland
Actions
Hi! I'm Aldan, and I'm passionate about technology and research. I studied Computer Science in Spain, France and Switzerland, founded a few associations and contributed to open source projects. I've also been lucky enough to get some recognition for leadership and academic achievements, and I love learning from the community and participating in events. Currently, I'm working on research that combines Knowledge Graphs and Natural Language Processing, both topics I'm always happy to talk about!
Links
Area of Expertise
Topics
Supercharge Your Coding with Fill-in-the-Middle Models: Beyond GitHub Copilot en
Dive into the world of fill-in-the-middle (FIM) models, the secret sauce behind tools like GitHub Copilot. We'll explore how these LLMs are technologically different from traditional models, and why they're a game-changer for code generation. But we won't stop there – come prepared to unlock their full potential!
You'll learn the details of FIM models, including tips for optimizing their attention architecture for superior results. We'll also go beyond code generation and discover alternative uses that can boost your productivity. And we'll look at some open source, privacy-friendly alternatives to GitHub Copilot that you can use with confidence in corporate environments.
Finally, we'll tackle its limitations head-on and discuss strategies for overcoming them. Ready to supercharge your productivity?
Google Summer of Code: How, Where, and Why? en
Want to get hands-on in software development and help the open-source community? Join me for an insight session on Google Summer of Code (GSoC). This annual program offers university students stipends to write code for open-source projects.
I'll share my experience contributing to the Django Software Foundation through GSoC. We'll look at how to apply, what to consider, and how to choose the right organization. You'll learn about the benefits of participating in GSoC, including professional growth, networking, and exposure to real-world development practices.
Cloud vs Local: Gemini Nano in Chrome en
Large Language Models don't have to live in the cloud... you can run them in your browser! 🤩
In this session, we'll look at one of the most exciting announcements from Google I/O: using Gemini Nano with the (experimental) Google Chrome built-in AI APIs for tasks like summarizing, rewriting, chatting, translating, and anything else you can think of! 🤔
We'll talk about the pros and cons – how this affects performance, privacy, data security and costs. We'll also learn how to configure Google Chrome to be able to call these APIs and use Gemini Nano 🤓
Ready to make the switch? Join us to explore the future of native AI! 🙌
Knowledge Graphs + LLMs: What's the latest in research? en es
Knowledge Graphs are super-flexible data structures that can capture complex relationships. Nice! But we can also put AI on top of them to get embeddings and start predicting things. Nicer! 🚀
Let's see what this is all about 👀 We'll learn about KGs, how traditional ML algorithms work on them, why they're limited, and how recent research advancements try to overcome that – using LLMs! 😉
Grafos de Conocimiento + LLMs: nuevas fronteras para la IA! en es
Los grafos de conocimiento son estructuras que nos ayudan a capturar y entender relaciones complejas entre diferentes conceptos. Son muy útiles para organizar grandes cantidades de datos diversos. Podemos entrenar modelos de inteligencia artificial para trabajar con estos grafos y descubrir nueva información, como cuando Facebook te sugiere amigos que podrías conocer.
Listo para explorar el mundo de la IA y los grafos de conocimiento? Aprenderemos cómo funcionan estos algoritmos, sus aplicaciones en la vida real y cómo superar sus limitaciones. Además, exploraremos brevemente las líneas de investigación actuales para combinarlos con modelos de lenguaje (LLM). Únete a nosotros para descubrir el futuro de la IA y cómo puede beneficiarte!
Give me the gist: JavaScript vs Lisp en
Did you know JavaScript has a distant brother? It does, and it shapes its very foundations. In this talk, we'll quickly explore the main commonalities between JavaScript and Lisp, to better understand what truly shapes the core of JavaScript. No previous experience required!
Attacking LLM Detectors with Homoglyph-Based Attacks en
As large language models (LLMs) become more and more skilled at writing human-like text, the ability to detect what they generate is critical. This session explores an attack vector, homoglyph-based attacks, that effectively bypasses state-of-the-art LLM detectors.
We'll begin by explaining the idea behind homoglyphs, characters that look similar but are encoded differently. You'll learn how these can be used to manipulate tokenization and evade detection systems. We'll cover the mechanisms of how homoglyphs alter text representation, discuss their impact on existing LLM detectors, and present a comprehensive evaluation of their effectiveness against various detection methods.
Join us for an engaging exploration of this emerging threat and to stay ahead of evolving evasion techniques!
Strolling Through the Garden... of Vertex AI en
In this session, we'll explore Google's AI models through Vertex AI's Model Garden. The Model Garden is a big collection of pre-trained models for many different uses, including natural language processing, computer vision, and structured data. It lets developers, data scientists, and AI fans use cutting-edge AI models without complex customization or a deep understanding of machine learning.
In this talk, I'll show you how to use different types of models from the Vertex AI Model Garden in your projects. We'll cover practical examples, showing you how to use models for text processing. By the end of this session, you'll understand how to use the Model Garden – and use it to speed up your AI projects!
How to use Gemini without getting caught en
Join us as we explore the world of detecting AI-generated text, such as what you get from Gemini, Claude or ChatGPT. Have you ever wondered how experts uncover it? We'll spill the beans!
We'll start from scratch, briefly explaining how the generation of text using AI works. No previous background in the field is required!
Then, we'll discuss classifier systems for detection, black and white box models, watermarking, and manual techniques for detection—cool stuff to keep in your toolbox!
But here's the twist—these systems are not foolproof. There are ways to get around them, and we'll explore how. Ready to uncover the mysteries? See you there!
Build your AI replica... with your WhatsApp chats and Google Cloud! en es
Have you ever been curious about how other people feel when they talk to you? Now you'll know!
In this session, we'll take our WhatsApp message history and use it to train our very own versions of Gemini (like ChatGPT) — it's going to be amazing! Your experiences, your opinions, your secrets... We'll use them all to build a chatbot that speaks and thinks like you. Your digital twin! 🙌
But be careful! Talking to yourself can be very addictive 😱 Please don't leave your friends and loved ones for your own AI! 😉
Join us! We'll keep things simple, no previous knowledge required!
Crea tu propio clon... con tus WhatsApps y Google Cloud! en es
Te has preguntado alguna vez cómo es hablar contigo? Anímate a descubrirlo!
En este taller usaremos nuestro historial de mensajes de WhatsApp para entrenar nuestras propias versiones de una IA. Tus experiencias, tus opiniones, tus secretos... cogeremos todo eso para construir un chatbot (estilo Gemini / ChatGPT / Claude / etc) que hable y piense como tú: tu gemelo digital! 🙌
Pero, cuidado - hablar contigo mismo puede ser muy adictivo 😱 No abandones a tus amigos y seres queridos por tu IA! 😉
How far are we from General AI, really? en
Short answer: very far.
If you just wanted the short answer, you can skip the session. But there's much more to it than that! 🤓
In this talk, we'll explore the current state of AI and its limitations. We'll discuss the hype around AI and why we're not as close to General AI as some might think. We'll also talk about some of the most promising avenues of research that (might) get us there in a very distant future.
Join us for a deep dive into the world of AI and a reality check on where we stand today. You'll leave with a better understanding of the field and a more critical view of the technology that surrounds us.
Are Spanish speakers more likely to ask chatbots like Gemini about sex? en
Did you know that there's public datasets of conversations of users with chatbots? We can learn a lot from them!
In this session, we'll explore the differences in the way that speakers of different European languages interact with chatbots. What are the most negative ones? Who's more likely to hate on others? Which ones are more satisfied with the answers they get? We'll analyze the data and interpret the results. You'll be surprised by what we find!
Gender bias and LLMs: why it's so hard to fix LLMs like Gemini en
We all know that AI can be biased. If we ask it to complete the sentence "My mum works as a...", we're likely to get "nurse" as a suggestion. But why is that? And how can we fix it?
Yes, data is a part of the equation, but this problem goes deeper. We'll learn why some bias is good (in specific cases). And we'll discuss ways to fix it, as well as its implications. Let's dive into the world of AI bias!
How on earth are we gonna enforce the AI Act? en
So – we've done it. The AI Act is a reality! Or is it?
Well... maybe not yet. In fact, there's a lot of open questions. In this talk, I want to help you understand the requirement that "the outputs of the AI system are marked in a machine-readable format and detectable as artificially generated or manipulated".
What does this mean? How can this be enforced? What are the implications for AI developers? There's no clear-cut answer yet (spoiler), but I'll give you my own two cents on the topic. Join us for a deep dive into the AI Act and its implications!
How private are your conversations with chatbots like Gemini? en
Talking to chatbots feels like talking to a friend, but are they really trustworthy? Well – not always!
In fact, your conversations may already be public on the internet. And I've found it really disturbing to see the kind of information that's out there. Social security numbers, addresses, phone numbers... you name it! 😱
In this session, we'll explore some of these conversations. We might have a few laughs, but my goal is that you'll walk out much more aware of how public all these conversations are. And you'll learn how to protect yourself and your data better. Join us!
NDC Security 2025 Sessionize Event Upcoming
DefCamp 2024 Sessionize Event Upcoming
Porto Tech Hub Conference 2024 Sessionize Event Upcoming
Devfest Milano 2024 Sessionize Event
Prompt Engineering Conference 2024 Sessionize Event
DevFest London 2024 Sessionize Event
DevFest Scotland Online
Attacking LLM Detectors with Homoglyph-Based Attacks
Manchester Tech Festival 2024 Sessionize Event
DevFest Bari 2024 Sessionize Event
DatafestAfrica 2024 Sessionize Event
NDC Porto 2024 Sessionize Event
DevFest Ponferrada 2024 Sessionize Event
Digital Rights 2024 Global Gathering
The event brings together digital rights networks from around the world for three days of collaboration building, knowledge sharing, brainstorming, and rejuvenation.
Nerdearla 2024 Sessionize Event
Data Makers Fest 2024 Sessionize Event
UGConf Perú 2024 Sessionize Event
Data2Day 2024
The Power of Knowledge Graphs and AI: A New Era for Developers and Data Scientists
Attack Surface Management Summit
Homoglyph-Based Attacks: Circumventing LLM Detectors
Salon Data 2024 Sessionize Event
The Azure and AI Show
Knowledge Graphs, AI & LLMs: new horizons for Developers and Data Scientists
Budapest Data+ML Forum 2024 Sessionize Event
DeveloperWeek Global 2024 Sessionize Event
Data Science Festival
Maker Faire Aarhus
What would a Maker Faire Bot look like? - Training an LLM with texts contributed by Maker Faire attendees
Getting the most out of ChatGPT: tips, tricks and ideas
Global AI Bootcamp 2024 - Lüdinghausen Sessionize Event
Conf42.com Large Language Models (LLM) 2024
How to use ChatGPT without getting caught
Global AI Bootcamp 2024 - Latam Sessionize Event
Global AI Bootcamp, Pune 2024 Sessionize Event
Global AI Bootcamp 2024, Philippines Sessionize Event
Global AI Bootcamp 2024 Sessionize Event
Global AI Bootcamp - London 2024 Sessionize Event
Global AI Bootcamp 2024 Sessionize Event
GDG UCD
Human or ChatGPT? Detecting AI-Generated Texts
Seminar on European Digital Rights and Cybersecurity
ChatMadariaga: on Digital Humans and their Ethical Implications
Your Digital Rights International Seminar
ChatMadariaga: A Digital Twin of Salvador de Madariaga
NerdNite Santiago de Compostela
How to use ChatGPT without getting caught
InnovAmes 2024
How to use ChatGPT without getting caught
2023 CESGA Conference
Artificial Intelligence for the Automatic Generation of Slides for Scientific Papers
DevFest TUD 2023
How to catch AI-generated texts
DevFest Santiago de Compostela 2023
President & Organizer
Artificial Intelligence Day
On the Detection of AI-Generated Texts
Artificial Intelligence Conference
On the Detection of AI-Generated Texts
European Youth Event
Questioned the Vice President of the European Parliament on the development of Artificial Intelligence in Europe to be up to speed with the rest of the world.
Maker Faire Galicia 2023
Human or replicant? How to catch ChatGPT
Pint of Science 2023
Human or replicant? How to catch ChatGPT
IAprendizaxe Forum
On the Detection of AI-Generated Texts
Python Web Conf 2022
Behind the scenes of Wagtail's search
DjangoCon US 2021
Full-text search (FTS) on your website - my Google Summer of Code project
Aldan Creo
Technology Research Specialist @ Accenture Labs
Dublin, Ireland
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