I treat a project as an end-to-end engineering chain: context, requirements, risks, architecture, implementation, testing, data feedback, delivery, and documentation.
Explore my public repositoriesClin · Project management + full-stack tooling
Turning complex systems into reliable, usable products.
I work across desktop applications, AI, automotive software, and engineering automation—solving real problems in reliability, efficiency, and data quality.
SoftwareDesktop apps & reliable systems
AISpeech recognition & data quality
AutomotiveSimulation, perception & control
AutomationTesting, CI/CD & toolchains
About me
I am Clin, focused on project delivery and full-stack engineering tools.
My work spans desktop software, speech AI, automotive systems, and automation. I care about turning requirements into maintainable systems whose failure modes and results can be verified.
Real environments first
Make trade-offs around hardware, network, endpoint, and deployment constraints.
Reliable data first
Build persistence, validation, fallback, and traceability into the main path.
Verification first
Use automated tests, quality gates, and field acceptance to prove results.
Featured projects
Engineering work across software, AI, and automotive systems.
Claims are based on current code, project documentation, or public repositories. Unverified metrics are intentionally omitted.
FEATURED CASE STUDYActive development
LanExamSystem
A LAN-based desktop examination system covering question banks, test publishing, answer recovery, centralized submissions, automatic grading, reporting, and archiving.
Engineering focusNetwork recovery · Server snapshots · Persistent grading queue
Software · AI
Resonance Music
A local-first Flutter music player with explainable, on-device recommendations.
AI · ASR
Sichuan–Chongqing Navigation ASR
An offline speech-recognition pipeline with anomaly detection, constrained retries, model fallback, and auditable output.
Automotive · Automation
CarMaker CameraRSI Recorder
An open-source multi-camera RSDS recorder with bilingual GUI, session manifests, and optional NVENC, QSV, or AMF encoding.
Engineering focusIndependent reconnect · Hardware encoding fallback · English / Chinese GUI
Automotive · Robotics
ROS 2 Racecar
Vision-based cone following with PID control, a mission state machine, and explicit fail-safe behavior.
Skills & expertise
Capabilities organized by engineering context.
Each group reflects technologies used or documented in current projects.
Software Engineering
C/C++, C#/.NET desktop applications, Python, MySQL, data structures, and algorithms.
AI / Data
Dialect ASR, offline inference, normalization, anomaly detection, quality gates, and auditable output.
Automotive
IPG CarMaker workflows, OpenDRIVE data, ROS 2, OpenCV, PID control, and Linux tooling.
Engineering
Project management, system design, automated testing, CI/CD, performance, reliability, and technical documentation.
Contact
Discuss a project, toolchain, or collaboration.
Please include the context, current problem, target environment, and desired outcome.