Dmytro Samoilov — IOS Engineer

On-device AI is the new performance budget. I build and document hybrid on-device + cloud architectures for Android apps. I'm a Senior Android Engineer with 10 years building production apps for 500K+ users in Kotlin. For the last cycle I've been going deep on a single, under-covered intersection: running AI on Android devices themselves — Gemma 4, Gemini Nano, AICore, ML Kit GenAI — and deciding, feature by feature, what stays on-device and what goes to the cloud. Why this niche? Nearly half of apps are expected to ship AI features in the next two years. Most mobile teams default to "send everything to the cloud." That's the expensive answer. The harder, more durable answer is hybrid: on-device where latency, cost, privacy and offline matter — cloud where scale and model size win. That trade-off is what I work on and what I write about. What I publish: • End-to-end builds: shipping AI features in real Android apps • Architecture breakdowns: hybrid on-device + cloud patterns, with benchmarks • Decision frameworks: when on-device AI actually pays off (and when it doesn't) • Tool reviews: Android Studio Agent Mode, Gemini Code Assist, the new tooling stack Who this is for: • Mobile leads deciding on-device vs cloud architecture • Teams that want fewer API bills, better latency, and offline-capable AI If you're building AI features in an Android app — or thinking about it — I'd like to learn from what you're seeing. DMs open.

Stackforce AI infers this person is a Mobile Application Development expert with a focus on AI integration and performance optimization.

Experience: 8 yrs 4 mos