Wenkai CheYou can also call me Calvin
I own large-scale mobile systems, ship AI-integrated products end-to-end, and carry a published machine-learning research record.
Senior Application Development Engineer · Beijing, China → United States (H2 2026)
About
A dual-track engineer.
Large-scale mobile ownership, AI-integrated product building, and a research foundation — in one person.
I'm a Senior Application Development Engineer at Tencent, where I'm the primary owner of the live-streaming module inside Tencent News — an Android client used by more than 30 million people. I unified personal, broadcast, and sports live formats under one architecture and drove releases across a team of 60+ engineers.
I build at the edges of what's new: I've integrated generative-AI features into the live product — an in-app AI agent chat assistant and an AI digital-human live-interaction experience — and, on my own time, designed and launched Case Monitor (pivib.com), a subscription web service that now serves 12.5K+ users on 100% organic growth, with a self-hosted vision model automating the status-reading pipeline end-to-end.
Underneath the product work is a research foundation: a master's in pattern recognition, 12 publications and 555+ citations in applied machine learning and hyperspectral imaging, and an active peer-review record across 8 international journals.
Experience
Seven years at Tencent.
Tencent News Android client — 30M+ users. Presented as dated, themed tracks.
Senior Application Development Engineer · Tencent
Beijing, China
Jul 2018 — Present
Live Streaming & Generative AI
2021 — PresentTencent News (30M+ users)
- Primary owner of the live-streaming module for the entire app — led its architecture, integration, and development; unified personal, broadcast, and sports live formats under a single system and coordinated cross-platform protocol design.
- Drove end-to-end release across a team of 60+ engineers.
- Integrated generative-AI features into the live product, contributing heavily to an in-app AI agent chat assistant and an AI digital-human live-interaction experience that brought AIGC conversation into the live-streaming flow.
Live streamingGenerative AIAIGCArchitectureCross-Platform Core — Kotlin Multiplatform
2024 — Present- Built shared business logic and Compose Multiplatform UI (qnCore / qnCompose / qnMedia) in a Kotlin Multiplatform codebase, unifying features across Android and iOS.
- Extended core modules to HarmonyOS.
Kotlin MultiplatformComposeAndroidiOSHarmonyOSRemote Build System
2020 — 2023Self-initiated · sole developer
- Independently conceived and built a zero-configuration remote compilation system end-to-end: an Android Studio / IntelliJ IDE plugin (Kotlin), server-side sync-and-build scripts over SSH & RSync, and Dockerized build servers.
- Reached 90%+ adoption across the QQ, Video, and News teams, cutting local CPU/memory usage ~90% with zero manual setup.
- Led a Git LFS migration that cut repository clone size from 11.95 GB to 500 MB — a 96% reduction.
IDE pluginDockerSSH/RSyncGit LFSDeveloper toolsAudio Playback
2019 — 2020Tencent News
- Designed and built the audio player library powering all in-app audio (channels, TTS narration).
- Eliminated loading delays with a local socket-proxy cache plus preloading for near-instant playback.
- Built a segmented TTS player with real-time, sentence-level highlighting.
ExoPlayerAudioCachingTTS
Projects
Things I've built end-to-end.
Independent work — designed, shipped, and operated solo.
Remote Build System
Zero-config remote compilation for 1000s of engineers
A self-initiated, single-developer system that moves Android builds off the laptop and onto Dockerized build servers — with no manual setup for the engineer.
- Android Studio / IntelliJ IDE plugin (Kotlin) plus server-side sync-and-build scripts over SSH & RSync.
- Led a Git LFS migration cutting repository clone size from 11.95 GB to 500 MB.
Skills
The toolkit.
Languages
- Java
- Kotlin
- Python
- JavaScript / TypeScript
Mobile
- Android SDK
- Kotlin Multiplatform (KMM)
- Jetpack Compose
- HarmonyOS
- ExoPlayer
- AAB / AAR
- Gradle
AI / ML
- Generative-AI integration (AI agents, AIGC)
- Local LLM deployment (Ollama, Qwen)
- RAG
- AI-assisted development (Claude Code)
- Machine learning
- Hyperspectral image analysis
Full-stack & Infra
- React
- Node
- Linux
- Docker
- VPS & self-hosting
- Cloudflare
- CI/CD
- Git
- Git LFS
Research & Peer Review
A published research record.
Applied machine learning and hyperspectral imaging for food-quality and safety detection.
Active peer reviewer for 8 international journals in spectroscopy, machine learning, and agriculture, including Scientific Reports, Journal of Food Science, Computers and Electronics in Agriculture, and Journal of Nondestructive Evaluation.
Education
M.S., Pattern Recognition & Intelligent Systems
Heilongjiang University2015 — 2018B.S., Computer Engineering
Yantai University2011 — 2015
Contact
Let's talk.
Open to Senior Android / Mobile, Senior Application Development Engineer, LLM / AI application engineer, and hyperspectral research roles in the U.S. — work-authorized, no sponsorship needed. The fastest way to reach me is email.