Android Sosom Meetup: How are developers at home and abroad utilizing AI?

Опис

[Important: RSVP Information] https://ticketa.co/event/5g8dos43 You must verify the purchase of your ticket on TicketaCo to gain proper access to the venue. If you are not registered on both platforms, your participation may be canceled. In an era where AI generates more code, how can we create better Android apps? Development using AI has started to evolve beyond simple code autocomplete and is changing the very way developers work. So, to what extent and how are real Android developers utilizing AI? During this Android Sosom Meetup, we will discuss various ways to apply AI in the development process based on the actual experiences of Android developers from home and abroad. 1. The Good Stuff, Not the Slop: Engineering High-Quality Android Apps with Modern AI Tooling Speaker: Daniele Bonaldo (Mobile consultant at Outfit Code, GDE Android) The rapid rise of AI coding assistants has led to a dangerous misconception: that anyone can effortlessly generate complex apps. The reality? Without strict architectural guardrails, pure "vibe coding" quickly degenerates into unmaintainable "slop." In this talk, we will explore the transition from simple AI autocomplete to true Agentic Development in Android Studio. We’ll see how to leverage our senior engineering experience alongside modern AI tooling to boost productivity without sacrificing quality. Expect a deep dive into practical strategies: enforcing deterministic architectural constraints using AGENTS.md and Android Skills, zeroing out context-switching with MCP, and validating features using natural language Journeys. Join me to discover how to guide AI agents to craft high-quality, user-centric Android applications where the developer remains the architect, not just a boilerplate writer. 2. The Delusion of Using AI Well, and How to Check It Speaker: Ha Dong-hyun (Remember) As AI coding tools improve rapidly, another delusion has emerged: the belief that simply producing more tools and increasing automation means one is using AI well. However, whether the results have improved due to my design or just because the model has improved cannot be known without separating and reassessing them. This presentation will address methods to doubt not the outputs created by AI but rather the tools I have created and the numbers I have generated. Based on failures encountered while creating search tools, server judgment tools, and design automation, we will examine the reasons unused tools are created, the process of fixing non-existent weaknesses despite diligent measurement, and points where documented rules are not followed. I will also share practical devices to avoid the traps of self-assessing one's tools—pre-registration, third-party testing, and sample extraction. This session is for those who have created multiple AI tools but are unsure if they have truly improved. Location: Google Startup Campus, Samseong Station Time: September 8 (Tuesday) 19:00 ~ 21:00