Embedded and industrial systems engineer based in Hefei, Anhui, China.
I build systems from hardware bring-up to application software, algorithms, and verification.
Email / 邮箱: clarkcui2you@gmail.com
我目前在安徽合肥，工作覆盖硬件、STM32/Linux 底层、应用架构、算法、测试平台与工程交付。我关注的不只是“功能做出来”，更重视能否通过 HIL、SIL、测量、回放和自动化测试证明它可靠地工作。
| Capability | Public proof |
|---|---|
| STM32 and flight-control engineering | Aerakia Flight Stack: drivers, RTOS, application integration, estimation algorithms, HIL, SIL, replay, and digital twin |
| Hardware-system development | dpiny RTK Base Station: isolated power, STM32 + GNSS, dual RS-422, firmware, diagnostics, and bench verification |
| AI-assisted engineering workflow | Liakia: repeatable build → flash → test → diagnose → evidence → handoff, proven on real STM32 hardware through dpiny |
| Linux and industrial product systems | Splicia Vision Platform: completed Linux camera/platform bring-up and diagnostics, plus a clearly marked Phase2 UI and modular architecture concept |
| Open hardware and protocol validation | SBUS-RS422 Bridge: isolated transparent hardware bridge, browser analyzer, logic-capture analysis, and input/output consistency testing |
A sanitized flight-control engineering portfolio showing the full stack on STM32H743: low-level peripherals and sensors, FreeRTOS services, application data flow, attitude-estimation research, HIL/SIL, digital-twin tools, and evidence-backed debugging.
Two views of one engineering result. dpiny demonstrates the hardware and firmware system; Liakia extracts the reusable AI-assisted delivery workflow used to build, flash, test, diagnose, document, and hand off that real system.
A privacy-safe, in-progress case study. Completed work covers embedded-Linux camera/platform bring-up, IMX296 topology and RAW-output diagnostics, and migration investigation. The published Phase2 UI and modular architecture are original design concepts; vision algorithms, motion control, closed-loop performance, and production UI are not claimed as completed.
An open-source hardware-and-software project that turns a small protocol bridge into a complete verification story: schematic, isolation, selectable polarity, live decoding, generic waveform import, consistency analysis, and a planned hardware-versus-MCU benchmark.
hardware / schematic / interfaces
STM32 / C / C++ / FreeRTOS / DMA / USB / Ethernet
Linux / camera pipelines / OpenCV / Python
serial protocols / GNSS / RTCM / SBUS / RS-422
HIL / SIL / digital twin / replay / automated evidence
JavaScript / TypeScript / Shell / PowerShell / engineering automation
understand → build → measure → diagnose → verify → document → share
I use AI to increase engineering throughput, while keeping hardware measurements, reproducible tests, source review, and human judgment as the acceptance boundary.
Most projects are active engineering work. Safety-critical use always requires independent review and validation for the intended environment.