26 au 28 février, 2025
Montréal, Canada
Conférence Intelligence Artificielle
La liste des présentations est sujette à changement sans préavis.
Intelligence Artificielle
Integrating AI into your application transforms the traditional software development and DevOps lifecycle. Unlike standard code, LLM outputs can vary for the same input, leading to different accuracy levels. This affects both pre-deployment testing and post-deployment monitoring. To achieve and maintain quality, you will learn to apply ML/LLM Ops practices from data science within your software development and DevOps lifecycles.
Intelligence Artificielle
Learn about advanced profanity detection patterns beyond simple keyword search. My session will cover data collection and training of a profanity detection model, and it's integration with live Speech-to-Text AI technologies. We will also look at the challenges around handling live data.
Intelligence Artificielle
In this session, we will explore the process of building single-purpose micro-agents and orchestrating them to form a versatile super agent. This approach leverages the strengths of each micro-agent, which possess their own planning, reflection, and self-critique capabilities, to collectively tackle complex tasks.
Intelligence Artificielle
Accessibility is more than just a checklist—it's a web fundamental. However, incorporating AI tools like GitHub Copilot into your workflow presents both new possibilities and unique challenges. In this case study, I’ll share my experience using GitHub Copilot to build accessible web components for Adobe Express. Join me to learn how to leverage AI tools like Copilot effectively, without losing sight of the critical role you play as a developer
Intelligence Artificielle
This talk explores AI's impact on software development, focusing on productivity enhancement. It covers AI applications like code completion, bug detection, refactoring, and test generation. Through demos and examples, attendees learn to use AI tools for various coding tasks across languages and environments. The session aims to equip developers with AI-driven strategies to boost efficiency and skills.
Intelligence Artificielle
AI coding promises a developer utopia, but the reality can feel like a wild west shootout. This talk ditches the hype and dives into the real-world challenges of building with AI. Discover the inner workings of popular prompt engineering techniques (Chain-of-Thought and In-Context-Learning), learn how to recover from unexpected AI behaviour, and gain the tools to wrangle the power of AI for your projects! Take a deep breath before attending.
Intelligence Artificielle
Unlock the power of Machine Learning by coding your own models from scratch! In this session, we’ll break down complex ML concepts into easy-to-understand steps. You'll start with a simple linear regression and move on to neural networks without relying on pre-trained models like GPT. By the end, you'll have the skills and confidence to implement Machine Learning in your projects. Join us to demystify ML and empower your ML journey!
Intelligence Artificielle
This session covers how embedding models can retrieve relevant method matches more effectively than traditional search. We’ll explore real-world examples, like migrating identity providers, where we can use embeddings to identify all HTTP authorization headers across libraries. Additionally, we’ll cover how the top-k selection technique improves search speed and accuracy, by leveraging code-as-data models which use Lossless Semantic Trees.
Intelligence Artificielle
Let's think a bit deeper and beyond the scandals with Cambridge Analytica, Facebook, VW- and see how our responsibilities go far beyond just delivering quality software adhering to requirements.
Moral implications, hard choices, ethical dilemmas with the blurred line between right and wrong - I'll share examples and try to make you think of what we can do in practice for leading an ethical professional life having a positive impact to society.
Moral implications, hard choices, ethical dilemmas with the blurred line between right and wrong - I'll share examples and try to make you think of what we can do in practice for leading an ethical professional life having a positive impact to society.
Intelligence Artificielle
Learn about advanced techniques in retrieval-augmented generation (RAG) going beyond simple vector databases. LLMs need guidance! My session will cover the spectrum from chunking and vectorization to sophisticated query transformations and the integration of agents, highlighting the role of prompt engineering and fine-tuning in optimizing RAG pipelines.
Intelligence Artificielle
Retrieval-Augmented Generation (RAG) is a technique that helps reduce LLM hallucinations by grounding responses in external, accurate data sources. However, taking a RAG beyond PoC is not easy. In this talk you'll learn about the challenges of naive RAG implementation, and what can help you build realistic RAG. We’ll cover topics like re-ranking, metadata filtering, BM25 and hybrid search, and evaluating RAG systems beyond a “vibe check”.
Intelligence Artificielle
Tired of generic chatbots? Unlock personalized AI shopping assistants with OpenAI's Function Calls API. Learn to build chatbots that tap into your product catalog, understand customer preferences, & deliver tailored recommendations, boosting engagement. From CEOs envisioning the future to developers building it, this talk empowers everyone to embrace AI-driven commerce.
Intelligence Artificielle
Comment utiliser l'IA générative sans tomber dans ses pièges ? Ce talk explore son rôle dans la production cinématographique, télévisuelle et marketing, en mettant en lumière réussites et échecs. Nous verrons comment elle peut aider à analyser des documents, valider des informations et automatiser des processus, tout en gardant le contrôle. Au-delà des solutions off the shelf, des exemples concrets montreront comment en tirer le meilleur parti.
Intelligence Artificielle
GitHub Copilot is everywhere. It’s on GitHub.com, in Visual Studio, Visual Studio Code, GitHub Codespaces and more. Did you know you can extend Copilot?
In this hands on talk, you’ll learn:
- the architecture of a Copilot extension
- how to leverage the Copilot extension preview SDK to build GitHub Copilot extensions
- how to build out your first Copilot extension
We’ll wrap things up by taking a peek at a real world example.
In this hands on talk, you’ll learn:
- the architecture of a Copilot extension
- how to leverage the Copilot extension preview SDK to build GitHub Copilot extensions
- how to build out your first Copilot extension
We’ll wrap things up by taking a peek at a real world example.
Intelligence Artificielle
Dans cette session, nous allons explorer comment mettre en place un pipeline CI/CD complet qui automatise le testing de bout en bout, le packaging et le versionnage des modèles, garantissant ainsi une transition fluide de la phase de développement à la production tout en maintenant un haut niveau de contrôle et de qualité.
Intelligence Artificielle
There's a deluge of new AI Agents coming online, especially in the realm of customer support. LLMs are powerful, but they are non-deterministic. How do we ensure that they are trustworthy, especially as they get plugged into APIs?
I'll dismantle the myth that AI agents replace human agents, and I'll discuss the principle of end user in control in bot design. I'll also talk about a new standard proposed by the IETF that will solve this problem.
I'll dismantle the myth that AI agents replace human agents, and I'll discuss the principle of end user in control in bot design. I'll also talk about a new standard proposed by the IETF that will solve this problem.
Intelligence Artificielle
Since the creation of my presentation, Postgres and the Artificial Intelligence Landscape, artificial intelligence use has exploded, with much anticipation about its future. This talk explores many of the advances that has fueled this explosion, including multi-dimensional vectors, text embeddings, semantic/vector search, transformers, generative AI, and Retrieval-Augmented Generation (rag). It then covers how databases and how the valuable data
Intelligence Artificielle
L’apprentissage par renforcement (Reinforcement Learning) est une approche de l'intelligence artificielle dans laquelle un agent apprend à prendre des décisions dans un environnement afin de maximiser une récompense cumulée. Le sujet aborderait les principes de base de cette approche, en s'appuyant sur des exemples pratiques avec Python et la librairie OpenAI Gym. Cette dernière offre une plateforme pour développer et tester des algo de RL
Intelligence Artificielle
RAG integrates external knowledge sources to improve performance by retrieving relevant documents during inference, making it ideal for tasks that require dynamic, up-to-date information. In contrast, Fine-tuning adapts pre-trained models to specific tasks through supervised learning, optimizing performance within static, well-defined domains. We analyze the trade-offs in terms of data requirements, deployment flexibility, cost, and use cases.
Intelligence Artificielle
By 2028, 75% of enterprise software engineers will use AI code assistants, transforming coding and agile development. Agile, created over 20 years ago, faces new challenges in the AI era. Do we need sprints when AI prioritizes tasks in real-time? What about cross-functional teams with virtual specialists available? Explore developers' evolving roles and the need to rethink collaboration strategies as AI becomes our coding partner.
Intelligence Artificielle
Have you ever encountered code having so many conditions and processing scenarios that it was impossible to maintain and extend t?
What about replacing it with a self-improving automatically generated algorithm?
With the rise of ML algorithms, artificial neural networks and LLMs, programming complex business rules and classification services can be successfully replaced with ML-based solutions. Let's see how to do it in your project!
What about replacing it with a self-improving automatically generated algorithm?
With the rise of ML algorithms, artificial neural networks and LLMs, programming complex business rules and classification services can be successfully replaced with ML-based solutions. Let's see how to do it in your project!
Intelligence Artificielle
In this presentation, you will see a case study illustrating the process of building an application based on RAG (Retrieval Augmented Generation) in PHP using a large text model (LLM) to efficiently find precise answers in a database of unstructured text data. I will also show how to use LLM in PHP on a local server and how to approach testing the generated results.
Intelligence Artificielle
GenAI is transforming the app landscape, but you don't need a PhD in AI to benefit! This talk goes beyond theory, revealing practical, "sneaky" ways to integrate GenAI into your existing projects. Learn to automate mundane tasks, personalize user experiences, and boost efficiency without disrupting your workflow. Real-world examples & engaging demos will inspire you to unlock the power of GenAI today!
Intelligence Artificielle
This presentation introduces AI basics for beginners, covering fundamental concepts and potential impacts. It then focuses on Spring AI, a tool simplifying AI integration in applications. The talk covers Spring AI setup, integration with Spring Boot, and implementation of common AI tasks like chatbots and sentiment analysis. Attendees will gain understanding of AI fundamentals and practical skills for incorporating AI into Spring applications.
Intelligence Artificielle
In this emerging AI era, delivering AI-based applications no longer requires data scientists. This shifts responsibility for AI output quality to QA and software engineers. Traditional tools and methods, however, are inadequate for measuring quality in these applications. Learn how to apply data science best practices to develop and maintain high-quality AI solutions.
Intelligence Artificielle
RAG is transforming information retrieval by integrating the pre-training of LLMs with your data, leading to novel insights and real opportunities for application features. This session aims to equip you with the knowledge to effectively implement RAG within your line-of-business apps. We'll focus on C# code samples, utilizing the Microsoft OpenAI Client to demonstrate building applications that will delight your users.
Intelligence Artificielle
Need the power of AI without compromising data privacy? Gemini Nano delivers! This compact, multi-modal AI model runs entirely offline – on your device or in the browser – enabling secure and private AI applications. Analyse sensitive financial data, process medical images, or transcribe confidential audio notes without sending data to the cloud. Build powerful Applications using Gemini Nano while staying off the network!
Intelligence Artificielle
Discover automated code remediation with the OpenRewrite refactoring engine, a deterministic, rule-based system. By manipulating the Lossless Semantic Tree (LST) representation of code, OpenRewrite ensures accurate, style-preserving transformations. We’ll demonstrate how AI enhances this process by performing impact analysis, searching through codebases, and assisting in refactoring, ultimately providing a scalable solution for modernization.
Intelligence Artificielle
Et si nous pouvions utiliser les LLM aussi simplement que n'importe quelle librairie de notre stack de développement préférée ?
Je vous propose de découvrir qu'il est désormais possible, simplement, d'intégrer les LLM au sein de vos applications.
Nous passerons par toutes les étapes afin que vous puissiez comprendre quelle approche vous convient le mieux.
Customiser un prompt, rajouter vos données, mettre des filtres, ...
Je vous propose de découvrir qu'il est désormais possible, simplement, d'intégrer les LLM au sein de vos applications.
Nous passerons par toutes les étapes afin que vous puissiez comprendre quelle approche vous convient le mieux.
Customiser un prompt, rajouter vos données, mettre des filtres, ...
Intelligence Artificielle
This talk will explore the cutting-edge advancements in computer vision as we enter 2025. We’ll delve into the role of multimodal large language models (LLMs) and their integration with visual data to enhance understanding and interaction. We’ll examine state-of-the-art object detection techniques like RT-DET and foundational models like SAM2, and explain the concept of grounding. Participants will gain insights into the latest trends.
Intelligence Artificielle
The rise of generative AI is reshaping the way we think about application development and deployment, allowing us to create smarter, more efficient solutions. But, the journey can feel overwhelming, from selecting the right foundational models, applying your unique data, and scaling across hybrid environments. However, the key to thriving in this emerging landscape lies in adopting the flexibility, transparency, and innovation of open source.
Intelligence Artificielle
I want to show you the secret weapon I have used for the past year. The GitHub Copilot lets me write full tests quickly, fills the gaps in my knowledge of 3rd party tools, and even writes clear descriptive commit messages. But it is not a "press the button to do it all". You need to guide the AI to do the right thing, which takes experience, but most importantly, you need to decide _what_ you want the AI to do, step by step. Remember: AI is
© 2010-2024 ConFoo. Tous droits réservés. Code de conduite
Gestion des Cookies
Pour offrir une bonne expérience, l'utilisation de cookies est nécessaire. Certains sont essentiels au bon fonctionnement du site, tandis que d'autres nous aident à mieux répondre à vos intérêts.