| Google Devfest Video | |
| Unleashing Gemini: Google Cloud's Generative AI Powerhouse. | Google keynote by Vijaykumar Jangamashetti (he / him) |
| The Next-Gen Smart Bot Revolution with Vertex AI Agent Builder: Powered by Conversational and Generative AI by Aniket Agrawal | Google keynote by Aniket Agrawal |
| Improving LLM Accuracy with Grounding by Shivank and Surabhi | Google keynote by Shivank and Surabhi |
| Break | |
| Empowering Communication: The quest for automatic solutions in sign language by Emmanuelle Aboaf | In an increasingly connected world, ensuring accessibility for all is essential. This talk explores the potential for automatic tools that bridge the communication gap for Deaf and hard-of-hearing people through the use of sign language. We will delve into the current challenges and advancements in developing AI-powered systems capable of understanding, translating, and generating sign language in real time. By examining the intersection of machine learning, computer vision, and linguistics, we aim to address the pivotal question: Can technology truly enhance accessibility without compromising the richness and complexity of sign languages? Join us as we explore the future of communication and inclusion through innovative solutions designed to empower those relying on sign language. |
| Accessibility First: Building an Inclusive App with Support for Various Disabilities by Enzo Conty | In this session, I will share my experience creating an app that prioritizes accessibility and supports different disabilities.
We will discuss the development of user interfaces that cater to a wide range of users, including those with visual impairments such as color blindness, cognitive disabilities, and mobility impairments.We will explore the techniques for designing inclusive UIs, such as high-contrast visuals, adaptable font sizes, and appropriate color schemes for color-blind users. Additionally, we will address the significance of semantic widgets to enhance accessibility.
Using practical examples, development insights, and fail attempts this talk aims to raise awareness about the importance of accessibility in apps. |
| Break | |
| Responsible AI: A Beginner's Perspective on Ethics and Fairness | As AI becomes more integral to our lives, creating responsible AI systems that are ethical and fair is crucial. However, navigating the complexities of responsible AI can be challenging, especially for those just starting in the field. In this session, I’ll share a beginner’s perspective on the foundational principles of ethical AI and the strategies emerging ML engineers can use to address fairness and responsibility in their projects.
We’ll explore topics such as bias detection, transparency, and the importance of inclusive data, along with practical steps for building more accountable models. Whether you're an experienced engineer or a newcomer to the field, this session will provide accessible, actionable insights into building AI that positively impacts society. Join me to discover how we can all contribute to a future where AI is both innovative and responsible! |
| Beyond Gadgets: How AI in Wearables is Reshaping Global Health by Jade Emmanuel | The potential of AI-powered wearable technology in healthcare is transformative, paving the way for a future where proactive, personalised health insights become accessible to all. In this session, I’ll introduce how AI is integrated into wearable devices to empower early diagnosis, improve chronic illness management, and reduce the burden on healthcare systems worldwide. We’ll break down the AI behind these devices—how it collects, analyses, and interprets data in real-time—and discuss the positive global impact that wearables can have in democratising healthcare.
Attendees will learn about AI’s critical role in wearables, including the technical processes that enable real-time health insights, and see compelling examples of how these devices are actively improving quality of life. We’ll explore the key trends driving the AI wearables market, including their application in telemedicine and remote monitoring, and discuss ethical considerations, such as data privacy and equitable access. Finally, I’ll offer a future vision of AI in wearables, emphasising how, with continued innovation and responsible development, these technologies can bring powerful health benefits to individuals everywhere.
This talk combines technical depth with real-world applications, making it both insightful for experienced developers and engaging for a broader tech audience interested in health and AI. |
| Break | |
| Level Up with AI: Empowering Developers of All Levels by Varun Kohli | Google keynote by Varun Kohli |
| The anatomy of a text-to-SQL chatbot by Dmitry Kostyuk | |
| Redefining Security: From Passwords to Biometrics and Digital Signatures by Vasilika Klimova | Are you tired of constantly forgetting passwords for various websites and applications? Or perhaps you're guilty of using the same password across multiple sites? If so, this talk is tailored just for you.
It's clear that passwords are becoming a relic of the past. You may have already encountered password-free sites. Many platforms are now embracing a new authentication protocol that allows you to access your accounts using biometrics or a special key.
But what if we take this technology a step further? Imagine using it to create digital signatures, facilitate agreements between parties, and even generate your own digital assets. During this talk, we'll explore how this technology functions and delve into its potential when integrated with blockchain technology. Don't miss out on this opportunity to be part of the conversation about the future of security and decentralization! |
| Let's build a modern camera app in Android by Tom Colvin | "Using the Jetpack CameraX library we can build modern camera experiences in Android apps super-easily.
This talk will demo how to integrate photo and video capture into your app, including:
- The latest updates to CameraX as seen in Google I/O '24
- The modern way to capture and preview images and video
- How to use device-specific features like portrait mode, night mode and face retouching
- Applying transformations such as rotations and cropping - or passing output to AI methods."
|
| From Android Wear to Gemini: The race to replace your phone | The concept of modern smart IoT devices has been evolving for over a decade, starting with the limited release of Google Glass in 2014 and expanding to smart clothing, watches, and fitness mirrors. Over the years, we’ve seen many ups and downs, from Virtual Reality to Augmented Reality. Now, we’re entering an era where Mixed Reality is the latest frontier, either worn on the body or embedded in our surroundings. Will smart goggles, smart pin and home appliances replace our phones? Why haven’t they achieved widespread adoption yet? In this presentation, I’ll explore the latest innovations in this space and discuss how advances in computing power and large language models are shaping the future of the industry. |
| Break Time | |
| Building Mobile Applications in the AI Era - Madona Wambua | Madona will delve into how artificial intelligence is reshaping the mobile landscape. She'll cover the various ways AI can be seamlessly integrated into your apps – from elevating user interfaces and delivering personalized recommendations to automating tasks and optimizing performance.
We'll examine real-world examples of AI-powered apps that are already changing the game and discuss best practices for ensuring ethical AI use and protecting user data. By the end of this talk, you'll be equipped with the knowledge and inspiration to build the next generation of intelligent mobile apps. |
| Mastering Data Flow in Modern Angular by Ankita Sood | Angular’s latest features revolutionize data flow, elevating the developer experience. Whether you are a seasoned Angular architect or a fresh-faced developer, this talk will equip you with the knowledge and best practices to make your components communicate in perfect harmony.
Angular is constantly evolving, and with it comes a powerful set of tools for managing data flow. Sharing data between components is a core challenge in Angular development. In this session, we’ll dive deep into the various mechanisms Angular offers for sharing data between components. We will delve into architectural patterns and best practices that ensure your data flow remains organized, efficient, and easy to maintain as your app grows. Key takeaways: - Comprehensive overview of data sharing options within angular components depending on their relation to one another. - Gain a deeper understanding of which data sharing option to choose in different scenarios. - Discover the differences between traditional and modern Angular data sharing methods, learning how to choose the most effective approach for each scenario. |
| Break | |
| Building High-Performance Computer Vision Systems with Vertex AI and Geminiby Dr Roushanak Rahmat | This technical talk will explore how to leverage Google Cloud's Vertex AI and Gemini to streamline and enhance computer vision workflows. We'll delve into practical techniques for data preparation, model training, deployment, and monitoring. Attendees will gain hands-on insights into real-world applications, such as object detection, image classification, and semantic segmentation. By the end of this session, you'll have the knowledge and tools to build and deploy robust computer vision solutions efficiently. |
| BigQuery vector search and multimodal embeddings by Marton Kodok | BigQuery allows you to generate multimodal embeddings and perform vector searches directly on your data without complex preprocessing steps. Simplify the process of finding relevant data, identifying patterns and trends, and clustering similar objects together. Learn how to generate embeddings using familiar BigQuery SQL syntax with multimodal inputs (text, images).
We’ll then demo how to use BigQuery’s vector search capabilities to explore data in new and innovative ways, leading to faster decision-making and improved insights. |
| Attacking LLM Detectors with Homoglyph-Based Attacks by Aldan Creo | 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 a novel 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!
|
| Tech & AI for Nature: a love story by Kalindi Fonda | In this session, we’ll explore real-world applications of AI in conservation, highlighting how the scale of tech can give us hope for our relationship with biodiversity, and why it is important that we use our skills for good, whatever that might mean.
I’ll share my journey in this field, insights from conversations with organizations, businesses, and individuals working with these technologies, and lessons from my own project.
Plus, we’ll review some innovative and fun efforts in the space: discover how photographing wildlife contributes to creating nature reserves, how satellites are protecting fish, and why drones sometimes carry wet napkins.
Can technology be the key to scaling up our efforts for meaningful environmental change? |
