Independent iOS Developer at Self-employed

Plan Flow Meter and AI Emoji Converter projects; built fully offline iOS apps with scheduling logic, iCloud/UserDefaults sync, local notifications, and an iOS widget system. Focused on privacy-first, backend-free architecture and end-to-end ownership from research to deployment.

Audio Machine Learning Algorithm Intern at Timeddomain Co. Ltd.

Collected and labeled 1,100+ hours of audio data for training voice models; utilized Whisper, OpenSpeaker, and Open-Unmix to improve dataset quality; contributed to targeted model optimization and personalized tuning for on-device deployment.

iOS Technical Director at Amazing Tech Co. Ltd.

Led the design, development, and optimization of the flagship iOS app XNote; conducted in-depth research on Apple PencilKit and CALayer rendering; implemented advanced drawing features and a local file management system; explored private APIs for extended native capabilities.

Senior iOS Developer & Serverless Developer at Pagoda Ltd.

Developed and optimized Pagoda to improve cold-start speed, UI responsiveness, network performance, and package size. Designed and maintained a serverless Node.js backend unifying API logic across Android, iOS, and web. Built App Clip with SwiftUI for offline promotions and refactored complex interfaces for performance.

iOS Developer at Tencent Technology Co. Ltd.

Contributed to an internal R&D team exploring innovative mobile music interaction models through continuous prototyping and experimentation. Developed multiple iOS demo apps focused on gesture-driven controls, beat quantization, multi-instrument playback, and audio compression using AVFoundation and AudioKit. Utilized advanced runtime techniques (e.g., method swizzling, associated objects) to build scalable components for behavior tracking and plugin-style architecture. Published representative components as open-source projects on GitHub: MikuLive2D (click to view on GitHub) – A gesture-controlled vocal app inspired by Mikutap, integrating Live2D avatars with beat-responsive animations and real-time lyric rendering. Designed visual effects and optimized buffer strategies for smooth performance and immersive feedback. UniversalPlayer (click to view on GitHub) – A lightweight audio player that supports both local and streaming media playback without third-party libraries. Built using low-level APIs (AudioFileStream, AudioQueue) and manually managed memory/pointer operations for compatibility with both 32-bit and 64-bit architectures.

Independent iOS Developer at Self-employed

Independent iOS developer owning end-to-end development for two App Store apps: Plan Flow Meter and AI Emoji Converter. Built fully offline apps with local storage, iCloud sync, widget support, on-device CoreML, privacy-first inference, and performance optimizations. Led research, design, development, deployment, and iteration for both products.

Audio Machine Learning Algorithm (Co-op) at Timedomain Co. Ltd.

Collected and labeled 1,100+ hours of audio data for training voice models. Used Whisper, OpenSpeaker, and Open-Unmix to improve dataset quality. Contributed to the GPT-SoVITS project with targeted model optimization and personalized tuning methodology.

iOS Technical Director at Amazing Tech Co. Ltd.

Led the design, development, and optimization of the flagship iOS app XNote. Researched Apple PencilKit and CALayer rendering; independently implemented advanced drawing features that significantly enhanced user experience and engagement. Developed a local file management system for flexible note sharing and explored private APIs to extend native capabilities.

Senior iOS Developer & Serverless Developer at Pagoda Ltd.

Developed and optimized Pagoda with a focus on cold-start speed, UI responsiveness, network performance, package size, and power efficiency. Designed and maintained a serverless Node.js backend unifying API logic across Android, iOS, and web. Built an App Clip using SwiftUI for instant access via QR codes, enhancing offline promotions. Refactored high-complexity interfaces for faster network requests and data processing efficiency.

iOS Developer at Tencent Technology Co.Ltd.

Contributed to an internal R&D team exploring innovative mobile music interaction models. Prototyped gesture-driven controls, beat quantization, multi-instrument playback, and audio compression using AVFoundation and AudioKit. Used runtime techniques (method swizzling, associated objects) to build scalable components for behavior tracking and plugin-style architecture. Published representative components as open-source projects: MikuLive2D and UniversalPlayer.

Independent iOS Developer at Self-employed

Led end-to-end development of two App Store apps as a sole developer: Plan Flow Meter (offline-first, SwiftUI, with widget support and iCloud + UserDefaults sync) and AI Emoji Converter (on-device CoreML-based emoji translation). Implemented local storage, offline scheduling logic, and performance optimizations while delivering a polished UI/UX and privacy-first design.

Co-op – Audio Machine Learning Algorithm at Timedomain Co. Ltd.

Collected and labeled 1,100+ hours of audio data for training voice models; enhanced datasets with Whisper, OpenSpeaker, and Open-Unmix; contributed to GPT-SoVITS project; developed personalized tuning methodology.

Senior iOS Developer & iOS Technical Director at Amazing Tech Co. Ltd.

Led design, development, and optimization of the flagship iOS app XNote; researched Apple PencilKit; implemented advanced drawing features; built a local file management system; explored private APIs to extend native functionality.

Mid iOS Developer & Serverless Developer at Pagoda Ltd.

Developed Pagoda app used by tens of millions of users; optimized cold-start speed, UI fluency, network performance, package size, and power; designed a Node.js-based serverless backend; created the App Clip; refactored complex service endpoints to reduce latency.

Mid iOS Developer at Shenzhen Dadi Yunwu Technology Co., Ltd.

Led development of an intelligent hotel check-in app; NFC-based room access and device pairing; validated end-to-end guest flow; embedded shared Flutter modules for cross-platform delivery.