Code ’25 Day 1 Session 3

With models like Gemini Nano running entirely on-device, the web is entering a new era of AI-powered applications that don’t require servers. This talk delves into the latest advancements in browser-based LLMs and tooling, including Gemini Nano, WebLLM, and ONNX.js. We’ll examine the trade-offs, performance considerations, and implications of these developments for AI on the web. Whether you’re building AI-powered applications or simply curious about the future of machine learning in the browser, this session provides a glimpse into what’s ahead.

Many of us are using AI assisted coding these days, however, many of us don't know how to leverage the full potential and what tasks we could use AI to assist with. In this session I will help you to unleash the full potential of GenAI to change the way you think when using AI assisted tooling like GitHub Copilot or CursorAI. This advanced session delves deep into how GenAI can elevate code quality, streamline testing processes, and bolster security measures. By integrating AI into your development lifecycle, you can achieve unprecedented levels of efficiency and robustness in your software projects. Key topics to be covered: Enhancing Code Quality: Discover how GenAI can assist in writing cleaner, more efficient code by providing intelligent suggestions, refactoring existing code, and ensuring adherence to best practices. Streamlining Testing: Explore AI-driven test generation, automated test case design, and continuous testing to reduce manual effort and increase coverage, leading to more reliable software. Bolstering Security: Learn how AI can identify vulnerabilities, suggest security improvements, and keep your codebase secure from potential threats through continuous monitoring and proactive defense mechanisms. Real-World Applications: Examine case studies and practical examples where GenAI has been successfully implemented to solve complex coding challenges and enhance overall software quality.

Did you know 1 in 5 Australians have a disability? That's 5.4 million people! Yet only 3% of Australian organisations are meeting accessibility standards. Can AI be the answer to making our digital assets more accessible? This talk explores the benefits and pitfalls of AI in achieving accessibility. We'll look at real world cases of how AI can be used to help accessibility initiatives, and areas where we need to be weary. How can we leverage this new technology to increase access? Hint: it's not what you think!