Staff Performance Engineer | Physics-Based Modeling & Algorithm Development
π PV Systems | π Automotive Safety | π Aerospace Engineering
Mechanical Engineer with 15+ years developing physics-based algorithms and computational models for:
- βοΈ Photovoltaic system performance modeling (current role at Maxeon Solar Technologies)
- π‘οΈ Automotive safety systems (10 years at ZF Active & Passive Safety)
βοΈ Aerospace control systems (graduate research at University of Michigan)
- 1M+ vehicles deployed with my production safety algorithms
- 500+ MW of solar capacity analyzed with my performance models
- 6 granted US patents in vehicle safety systems
- 6 publications with 30+ citations
- 2 open-source packages for PV Performance Modeling research community
Libraries & Frameworks
NumPy | Pandas | Scikit-learn | PVLib | Shapely | Matplotlib | Plotly
Tools & Platforms
Git/GitHub | Simulink | Dash | LaTeX | CATIA V5 | JMP
Core Competencies
Multi-physics modeling | Algorithm development | Statistical analysis | Machine learning | Optimization | Sensor fusion
βοΈ v_PVMismatch
Vectorized version of SunPower's PVMismatch using dynamic programming
- Physics-based electrical shade model for PV systems
- Integrated with Shapely for 3D geometric algorithms
- Tech: Python, NumPy, Shapely
πΊοΈ Tiger Conservation India
ArcGIS Story Map analyzing tiger population recovery patterns
- Spatial and temporal analysis of conservation success (2006-2022)
- GBIF biodiversity data + WDPA protected areas
- Tech: Python, ArcGIS, GeoPandas
β‘ PVShadeSim
Physical and electrical shade model using Shapely and PVMismatch
- Validated across 200+ field installations (RΒ² > 0.95)
- 10x performance optimization using caching
- Tech: Python, PVLib, Shapely
- π PV Physics-based Performance models.
- π Machine learning and optimization applications in renewable energy
- π± Wildlife conservation & GIS analysis
- πΌ LinkedIn
- π§ kbsub@umich.edu
- π Google Scholar
- π San Francisco Bay Area


