On-Device AI for Android Apps

I help Android teams ship AI features that work on-device, reduce API costs, and don't drain battery.

See how

About

Dmytro Samoilov

I've spent 10 years building Android apps — 20+ projects, 500K+ users, the full stack from architecture to performance.

Now I'm focused on the next shift: on-device AI.

Gemini Nano, Gemma, hybrid cloud/on-device architectures — I test these on real devices, benchmark what actually works, and document everything. Not theory. Real code, real numbers.

I also share what I learn on YouTube and LinkedIn, so developers and engineering teams can make informed decisions about AI on Android.

Based in Lisbon.

Work with me

I'm currently taking on select projects around on-device AI for Android — if that's relevant to your team, let's talk.

Get in touch

What I'm building

Claude Code Template for Android

Open-source template for using Claude Code with Android projects. Configs, prompts, and workflow automation.

View on GitHub →

On-Device Image Generation (iOS)

iOS app using Apple Intelligence to generate images locally. No cloud API. No data leaves the device.

View on App Store →

Hybrid AI Android App in progress

Building a demo app that compares pure cloud vs hybrid on-device/cloud AI architecture. Real benchmarks on real devices. Coming soon.

Follow progress on YouTube →

Blog

Case studies, field guides, and notes from shipping on-device AI on Android.

Guide

Local LLMs on Android in 2026: A field guide for product owners

The library landscape is churning, the device fragmentation is real, and the privacy story is bigger than the cost story. A field guide for the person making the call — written for CEOs, PMs, and engineering leads.

Read the guide →
Case study

Hybrid AI Cost Savings on Android

Real numbers from three production scenarios on a Pixel 8 — how moving voice cleanup, receipt parsing, and email drafting from cloud-only to a hybrid on-device + cloud architecture cuts AI costs by ~76%.

Read the case study →

Learn with me

I test on-device AI tools on real Android projects and share what I find.