Amorn Apichattanakul
I ship privacy-first AI — face liveness eKYC, biometric authentication — inside a financial super-app serving 4M+ users.
Staff Mobile Engineer at KBTG · Flutter GDE · Deep learning since 2018
16 years in mobile
iOS · Android · Flutter
Staff Mobile Engineer @ KBTG
Since 2019 at KBTG
Face liveness eKYC · 4M+ users
Security · compliance · zero downtime
I've been a mobile engineer for 16 years, since the iPhone 3GS. I began studying deep learning in 2018 (Deep Learning Specialization, Stanford/Coursera) — well before the on-device AI wave. And since 2019 I've built inside the hardest arena I could find: financial banking at KBTG, where I serve as Head of the Mobile Guild (60+ engineers) and lead a team of ~10 mobile engineers in a 3,000-person tech organization. Those three tracks now converge in production: LiteRT face liveness eKYC serving 4M+ users.
My work sits where mobile, AI, and banking meet — not as a job pivot, but as a deliberate, multi-year alignment. I lead by setting architectural standards (Flutter-Native hybrid strategy, security, observability) while staying hands-on with the code that proves them. Three principles guide that work:
This is the work I've been quietly preparing for. If it intersects with what you're building, I'm always up for a conversation.
On-device AI is easy to demo and hard to ship. Banking is where the constraints are real.
On-device LiteRT face liveness confirms a real human — not a photo, video, or mask — entirely on-device, with zero network round-trips during detection. Only after liveness passes does the verified frame go to the server for face comparison against the ID photo.
I designed a high-performance Flutter eKYC approach and published it — other banks have since adopted it in their own apps.
How I built it — Medium ↗Anomaly-triggered enforcement policies, zero security breaches, and a flawless compliance record at 4M+ user scale.
A zero-downtime strategy that absorbed major platform shifts (Privacy Manifests, OS updates) — governed for 60+ engineers.
Regulated eKYC onboarding demands real-time face liveness — confirming a real human, not a photo, video, or mask — running entirely on-device with zero network latency. Once liveness passes, the verified frame is sent for server-side face comparison against the ID photo. Since 2022, I've led the integration and ongoing optimization of LiteRT face liveness: architecting iOS/Android bridges via platform channels, tuning inference across device variance, and coordinating across data science, mobile, and security teams.
Result: production on-device liveness detection running across millions of daily authentications. I published the high-performance Flutter approach behind it (read on Medium) — other banks have since adopted it in their own apps.
A financial app serving millions needed enterprise-grade performance under strict transaction security and volatile network conditions across diverse devices. I architected a high-performance optimistic-loading architecture — executing flows instantly while resilient fallback callbacks catch transactional anomalies and securely enforce safety policies (logging the user out on critical mismatches). I also directed the optimization of biometric security loops by eliminating redundant platform-channel context handshakes.
Result: the fastest biometric login speed in the financial industry, with a flawless compliance record and zero security breaches.
Scaling mobile architecture standards for 60+ engineers within a 3,000-person tech organization, while keeping a financial super-app stable for a massive user base. As Head of the Mobile Guild, I define the Flutter-Native Hybrid Strategy and core technical standards, act as the primary technical authority on complex architectural decisions, and own the CI/CD and observability infrastructure (Firebase, Analytics) that lets the org move fast safely.
Result: standardized coding practices, reduced technical debt, and a "Zero Downtime" strategy that absorbed major platform shifts (e.g., Privacy Manifests) without service interruption.
Where the next wave of on-device intelligence is heading — and what it will mean for banking.
On-device LLMs for cross-platform Flutter apps — Gemini Nano, MediaPipe LLM Inference, and lightweight open models. Personalized assistants without server dependencies, with privacy preserved on the device. Exploring quantization, prompt engineering for small models, and integration paths.
Flutter & Flash: AI Mascot on the Edge — talk ↗AI-driven UI generation in Flutter. Streaming widgets from agents instead of rendering static screens. A new paradigm for adaptive interfaces.
Pushing LiteRT performance further — quantization-aware training, hardware-accelerated inference (Metal / NNAPI), and model-size reduction for production constraints.
Disseminating production-scale mobile and AI engineering insights to the global developer community.
Exploring on-device AI as a complement to cloud AI. While powerful models live in the cloud, many personal and responsive features benefit from running directly on the device — this talk covers practical implementation using TensorFlow Lite (LiteRT).
TALK
Feb 18, 2026
Exploring Agent-to-UI (A2UI) architectures using dynamic Generative UI (GenUI) in Flutter. AI agents that stream, construct, and render customized UI widgets in real-time.
View Slides → Watch on YouTube →
TALK
Nov 23, 2025
On-device AI using Gemini Nano to create an AI-powered mascot on edge devices. Flutter integration with on-device AI for personalized experiences without server dependencies.
View Slides →
TALK
Jul 26, 2025
Why choose between native and Flutter when you can use both? Flutter's Add-to-App feature for seamlessly integrating Flutter modules into existing iOS and Android projects.
View Slides →Mar 24, 2026
In 2023, when Gemini was first released, we were amazed AI could answer questions we couldn't even find on Google. By 2026, we've grown tired of just reading.
Read on Medium →Dec 3, 2025
Have you ever looked at your app's mascot, that friendly face guiding your users, and thought, "wouldn't it be amazing if it could… change?"
Read on Medium →Dec 15, 2023
Continuing my previous article on Face Liveness Detection in Flutter — enhancing performance with a native approach instead of Flutter workarounds for image transfer.
Read on Medium →Open to conversations about mobile, on-device AI, and banking-grade engineering.