Research-oriented builder at the intersection of Bioinformatics · Machine Learning · Practical AI Systems
🎓 Background in Electronic Information Engineering
I focus on turning research ideas into executable systems, especially in:
protein & peptide modeling · biological interaction prediction · graph learning · reproducible ML workflows · cross-domain engineering systems
I work on problems that require both theoretical understanding and practical implementation.
My interests sit at the intersection of:
- 🧬 Biology × Artificial Intelligence
- 📊 Structured Data × Representation Learning
- 🧪 Research Ideas × Engineering Execution
- 🔬 Scientific Modeling × Reproducible Workflows
- ⚙️ Electronics × Intelligent Systems
I enjoy breaking down complex problems and turning abstract ideas into systems that actually run.
- Protein / peptide property prediction
- Protein–protein interaction (PPI) prediction
- Cold-start biological relation inference
- Multimodal biological data fusion
- Knowledge-guided learning
- Graph representation learning
- Equivariant & diffusion-style modeling
- PyTorch-based training pipelines
- Experiment design & ablation studies
- Data preprocessing workflows
- Structured logging & result analysis
- Reproducible research pipelines
- Structured data processing (Pandas)
- Statistical reasoning & interpretation
- Scientific visualization
- Geographic / panel / DID-style modeling
- Circuit simulation: Multisim · Proteus
- PCB design: Altium Designer
- Embedded systems: Arduino · 8051 · STM32F
- FPGA / industrial tools: Quartus · GX Develop
- Integer / Nonlinear Programming
- Constraint optimization
- Global optimization algorithms
- Real-world applied optimization
- Game development with Unity · Godot
Languages Python · C · MATLAB · SQL · Markdown
AI / Data PyTorch · PyTorch Lightning · Pandas · NumPy · Scikit-learn
Engineering Multisim · Proteus · Altium Designer · Quartus
Tools VS Code · PyCharm · AutoDL · Git · Jupyter
- Knowledge-guided multimodal frameworks for cold-start PPI prediction
- Research workflows for scientific modeling
- Practical automation tools with Python
- Systems connecting biological data · modeling · interpretability
- Prefer clear frameworks over vague ideas
- Focus on usefulness, structure, and execution
- Turn messy problems into organized systems
- Value depth over surface-level knowledge
- Optimize workflows after understanding the full system
- Move smoothly between research thinking ↔ engineering implementation
Build useful things. Understand first principles. Prefer substance over appearance. Iterate continuously.
Open to collaboration, research discussion, and interesting technical projects.