Articles

Writing on mobile, AI & security

Practical notes on building AI-powered React Native apps, mobile architecture, secure application design, and leading a production mobile platform. No tutorials — just engineering analysis.

React NativeMobile ArchitectureAI Mobile AppsMobile SecuritySecure AIBackend SystemsEngineering Leadership

18 posts

All writing

Mobile Security

Why Mobile Security Should Return Risk, Not Booleans

isRooted() returns true — now what? A boolean can't answer the question your app actually has: given everything I can observe, is it safe to do this right now? The case for risk-based, explainable security on mobile.

Jun 202610 min read

Mobile Architecture

Building an Offline-First Media Playback Engine

Playing a file is easy. Keeping audio alive across flaky networks, app backgrounding, and OS process kills is the real engineering. How I rewrote a monolithic player into ~15 single-responsibility modules.

Jun 202612 min read

AI Mobile Apps

Building AI-Powered Mobile Apps Without Sacrificing UX

Bolting an LLM onto a mobile app is easy. Making it feel fast, native, and trustworthy is the hard part. Streaming, optimistic UI, offline fallbacks, and where the work actually belongs.

Jun 202610 min read

Mobile Security

Mobile Security Considerations for AI-Powered Apps

AI features expand a mobile app's attack surface in ways traditional checklists miss. Where keys live, what leaves the device, prompt injection on mobile, and backend-authoritative trust.

Jun 202611 min read

Engineering Leadership

Lessons From Leading a Production React Native Platform

What changes when a React Native app stops being a project and becomes a platform: release discipline, crash budgets, native boundaries, and the human side of mobile technical leadership.

May 20269 min read

React Native

Surviving a Major React Native Framework Upgrade

Upgrading the framework, UI library, and JS runtime on a mature app with a large native-module surface — without freezing releases. How to sequence, stage, and validate a foundational bump module-by-module.

Jun 202610 min read

Mobile Architecture

Background Audio Is Where Mobile Gets Hard

Foreground/background transitions, lock-screen controls, connectivity changes, and the OS killing your process — all different on iOS and Android. The platform edge cases that make background media reliability the real work.

Jun 20269 min read

Mobile Architecture

How AI Changes Mobile Product Architecture

Adding intelligence to a mobile product reshapes its architecture — proxy layers, context assembly on-device, streaming state, and the new failure modes you have to design around.

May 20268 min read

AI Mobile Apps

Designing AI Features That Feel Native on Mobile

The difference between an AI feature that feels bolted-on and one that feels native is mostly engineering — latency, motion, and respecting platform conventions.

May 20267 min read

React Native

Why React Native Engineers Should Understand AI Systems

Mobile engineers don't need to train models — but understanding how AI systems behave makes you dramatically better at building the products that ship them.

Apr 20266 min read

AI Mobile Apps

What Mobile Engineers Need to Know About RAG

Retrieval-augmented generation from a mobile engineer's seat: what to build on-device, what to push to the backend, and how to keep the experience fast.

Apr 20268 min read

Mobile Security

Secure Authentication Patterns for Mobile Apps

Token storage, refresh-and-retry, biometric gates, and session expiry done right — practical authentication patterns for production iOS and Android apps.

Mar 20269 min read

Mobile Architecture

Why Mobile Performance Still Matters in AI Products

AI is not an excuse for jank. Keeping the JS thread free, virtualizing lists, and offloading to native are what make an intelligent app actually feel good to use.

Feb 20267 min read

Secure AI

The MCP Security Problem: AI's New Supply Chain Crisis

MCP is rapidly becoming the connective layer for modern AI agents. It also introduces an entirely new class of security risks — from prompt injection to tool poisoning to supply chain compromise.

May 202610 min read

Secure AI

Security Considerations for AI Agents

Autonomous agents expand the attack surface in ways that traditional security models weren't built for. Prompt injection, tool misuse, and exfiltration risks need first-class treatment.

Feb 202613 min read

Backend Systems

The Hidden Complexity Behind Production AI

Most teams underestimate what it takes to run AI reliably at scale. Latency budgets, fallback chains, cost controls, and observability are non-negotiable.

Apr 202611 min read

Backend Systems

Observability for AI-Powered Products

Metrics, traces, and logs tell half the story. Instrumenting LLM pipelines requires a new class of signals: semantic drift, token budgets, and prompt telemetry.

Mar 20269 min read

Backend Systems

Why AI Systems Are Becoming Infrastructure

The shift from AI as a feature to AI as infrastructure is happening faster than most engineering orgs are ready for. Here's what that means architecturally.

May 20268 min read