Just got out of @zsmb.co Droidcon talk on Compose's Hot Reload, and I'm really impressed. It now looks and feels like a solid product. Also, super interesting to see some of the work that happens under the hood to make it seamless. Shows a lot of the platform's commitment to DEVEX.

Has anyone ever heard of tombstone files? I can't believe I never heard about this before after so many years. Also, I wish I didn't.

Gmail MCP is OP Been using it for one day and already hooked

Still buzzing from my time at Droidcon Lisbon! Thanks to everyone who came to my talk and for the great conversations and feedback. Massive thanks to Bruno and @cafonsomota.bsky.social and all the organizers. Fantastic conference! Slides are here: iurysouza.dev/speaking/dro...

🥹🥹🥹

Is this a test?

Made it to Lisbon! But as we took off at 7am I had a brilliant idea: Why not add another 30 slides to make a point? Improv time! See you at 14:10 on track 1!

I was supposed to arrive in Lisbon yesterday, but missed a connection due to a delayed flight. Leaving Paris now. Long day ahead 😅

I'm happy to share I'll be speaking at Droidcon Lisbon next month! If you're interested about on-device Gen-AI you'll like this one. The plan is to dig into how LLMs actually work and bridge to the tools that enable on‑device inference. Also down for a pastel de nata + coffee if you’re around 🇵🇹

Spending some time in Brazil with a proper home office setup 😌

Seems to be a lot of talk about Haze and Liquid Glass… At its core, Haze is a background content transform library, and the ‘visual effect’ (blurring, etc) is just an implementation detail. Liquid Glass will be a runtime shader like others which Haze already uses.

Apple sitting out of the whole AI assistant wars is nuts.

machinelearning.apple.com/research/int... Apple's new on-device solution is kinda like Google's AI Edge SDK, but as usual, less fragmented. Now, WTH are they comparing their model to Gemma 1 when Gemma 3 is out since March? I'm guessing their model is still quite basic. Lets see.

yes, but

Wireless debugging is actually happening this time!

AS new Agent Mode can understand Gradle structure and use IDE features. This will really set AS apart from alternatives which have no idea about kotlin/gradle context. Really good to see investment on that front! (There's also a ton more stuff. Huge release!) www.youtube.com/watch?v=KXKP...

Union types + Rich Errors Still not clear to me how union types will work (typescript PSTD), but Im all in for rich errors

Bold move 👀 github.com/Kotlin/kotli...

Google I/O update: New Gemma 3n, a 4B* model getting close to Sonnet 3.7 is both mindblowing and inevitable. Edge AI++

This was released less than 6months ago

New blog post up on UI layer architecture for persistent UI elements like navigation bars, navigation rails, and floating action buttons! www.tunjid.com/articles/ui-...

PSA: The Android Show (pre-io android only event) happens today! www.youtube.com/live/l3yDd3C...

Life update: 1h35m HM PB let's goooo

Apps exposing features via App Functions is the biggest change nobody’s talking about. MCP, but for android. developer.android.com/reference/an... Coming out on Android 16! Looking forward to this announcement on Google IO next month. Local Models + App Functions in the span of a year. Nice.

blog.jetbrains.com/idea/2025/04... This is a great move!

Gemini 2.5 is the real deal! Great move to add it to Firebender, guys!

The video is out! Thanks @louis993546.com for recording and sharing it! I changed the talk on short notice, so this is actually: "Putting The Genie in the Bottle A Crash Course on running LLMs on Android" Check it out: www.youtube.com/watch?v=VvJ1...

The "S" in #MCP stands for Security, by Elena Cross elenacross7.medium.com/%EF%B8%8F-th...

🪿🪿🪿

WOW 🤯 Gemini 2.5 Pro (SOTA model) is now available to all users for FREE on the web. If you haven't used Gemini cuz it didn't perform well in the past, it is time to try again.

Gave a talk yestersay using this. Pros: Worked pretty well. Whole content sitting inside my vault. Just markdown text. Cons: CSS

Shout out to the @berlindroid.de GDG! Thanks for having me 😄

💀 www.smbc-comics.com/comic/punish...

@jetbrains.com has joined the chat. Tried Junie's EAP version with my github.com/iurysza/modu... project. IMO the sweet spot for these kind of tools are projects with a clear (and narrow) scope. Worked super well so far. I liked how it was smart enough to figure out how to run tests by itself.

After 2 firmware updates and flashing 2 OS versions, it's alive!

What to expect during Android Makers by droidcon 2025? We just published an article to pump you up! medium.com/android-mak...

Introducing Gemma 3. The most capable model you can run on a single GPU. Cloud Run offers 1 GPU per instance, it is a perfect fit. Deploy it in one simple command. Blog: cloud.google.com/blog/product... Tutorial: cloud.google.com/run/docs/tut...

Got this while attending the Gemma Dev Day last week. Let's see if I can get it to do something cool now =D From I got so far, it can be great for hacking about. For LLMs it means models under 8B params. My hunch is it probably has the "AI performance" a phone will have in a couple of years.

www.youtube.com/watch?v=jwev... Kotlin 🤝 MCP @sebi.io nailed it on this one! I rly want to try this out! github.com/SebastianAig...

youtu.be/cMApO0UhqEs Have to agree with @t3.gg on this. The amount of AI generated code is increasing so fast, that yes, in a few months 90% of all code will be AI generated. There's a tsunami of vibe coders already flooding github. It's not stopping.

Wrapping up with Gemma Dev Day and a solid lineup. I have a hunch/hope that we're about to see a replay of mainframe vs microcomputers debacle. Only this time, it's the open model running on the edge vs the large proprietary ones. Gemma3 is definitely moving that needle 🔥

I'm in Paris 🇫🇷 today for the Gemma 3 release. Small language models are getting super capable and I'm excited to see this running on a device soon! Official tech report: storage.googleapis.com/deepmind-med...

Let's go!

Anyone in Paris next week? 🇫🇷

If you missed the #AndroidXR session from @mariobodemann.bsky.social last week, you can catch up on our YouTube channel! www.youtube.com/watch?v=RsFL...

firebender.com This is basically cursor for Android Studio/IDEA. Similar UX and even keymaps. You can use your own API keys (although you can use a lot of models for free??) and even supports local LLMs. Has an `agent` mode too! YC backed, and free (for now at least). Looks promising!

:cheffkiss: @scottwlaschin.bsky.social

Not sure how Raycast's implementation looks like under the hood given that they had to roll out their own Ray-1 model to interact with extensions, but it does feel like a somewhat opinionated take on MCP. bsky.app/profile/rayc... Now, hear me out: What if our Apps could provide MCP servers?

Been trying to find a decent alternative for TTS. This is what I got so far. TLDR; OSS is pretty close: www.zyphra.com/post/beta-re...

I keep finding SharedPrefs in libraries and codebases. If You're using it to store somewhat important stuff you may want to reconsider. Interesting thread github.com/flutter/flut... Official blog post: android-developers.googleblog.com/2020/09/pref...