About Me

I'm a Software Engineer and researcher based in Orange County, CA. My work focuses on robotics perception, 3D vision, spatial reasoning, and ML systems, alongside research in inverse photonic design using residual tandem neural networks.

Experience

Software Engineer II - Kazvu Labs

November 2025 - Present | Irvine, CA

• Design and develop end-to-end robotics perception pipelines using ROS 2, RGB-D sensors, LiDAR, Open3D, and PCL for real-time and offline 3D scene understanding.
• Build systems for point cloud generation, segmentation, classification, target point generation, geometric reconstruction, and spatial reasoning from multi-sensor perception data.
• Implement ICP-based registration, TF2 coordinate transforms, and multi-view geometric fusion workflows for scene alignment, spatial registration, and reconstruction.
• Develop volumetric mapping and safety-oriented perception systems using voxelized occupancy reasoning, TSDF/NVBlox concepts, intrusion detection, and online scene understanding.
• Build reusable C++ and Python perception libraries and integrate them into ROS 2 robotics stacks with sensor ingestion, transform-tree management, and real-time streaming workflows.

Machine Learning Engineer II - Hansji Corporation

June 2025 - November 2025 | Irvine, CA

• Developed predictive models and decision algorithms for booking pace forecasting, dynamic pricing, pricing elasticity, and revenue optimization across a multi-property portfolio.
• Built the Revenue Management System into a production ML platform through training, validation, deployment, monitoring, and scalable experimentation workflows.
• Designed MLOps infrastructure and GCP integrations supporting automated retraining, near real-time rate updates, and production pricing automation.
• Partnered with BI analysts and leadership on benchmarking, A/B testing, and translation of model outputs into business decisions.

Machine Learning Engineer I - Hansji Corporation

September 2024 - June 2025 | Anaheim, CA

• Built predictive models for dynamic pricing, booking pace forecasting, and revenue optimization across multiple properties.
• Engineered data pipelines for ingestion, feature engineering, model training, and validation to support scalable model development.
• Supported operationalization of RMS components with GCP services and near real-time rate update workflows.
• Worked with BI analysts on model evaluation, feature analysis, and A/B testing to measure business impact.

Machine Learning Engineer Intern - Hansji Corporation

June 2024 - August 2024 | Anaheim, CA

• Developed early machine learning and algorithmic prototypes that became the foundation of the Revenue Management System.
• Built initial data processing and feature engineering pipelines for multi-source property data.
• Tested time-series, regression ensemble, and rule-based optimization approaches for pricing-adjustment logic.
• Created proof-of-concept tooling for model evaluation, hyperparameter tuning, and baseline performance tracking.

Graduate Researcher - Estakhri Lab, Chapman University

Aug 2023 - Aug 2025 | Orange, CA

• Researched inverse design problems in photonics using residual tandem neural networks (forward + inverse models).
• Implemented residual blocks to stabilize training and improve accuracy for ill-posed inverse mappings.
• Built data pipelines and training/validation workflows; tuned hyperparameters and evaluated generalization under noise.
• Published peer-reviewed papers; contributed methods, experiments, and writing.

Skills

Publications

• Estakhri, N.M., Zahraii, P., Kashanchi, S., Estakhri, N.M. (2024). "Predictive residual neural networks for optical trapping of small particles." Optical Trapping and Optical Micromanipulation XXI, SPIE, Vol. 13112, pp. 111-116.
• Zahraii, P., Kashanchi, S., Estakhri, N.M., Estakhri, N.M. (2025). "Machine learning-driven prediction and inverse design of optical forces near gradient metasurfaces." Optical Trapping and Optical Micromanipulation XXII, SPIE, PC135800J.
• Zahraii, P., Kashanchi, S., Estakhri, N.M., Estakhri, N.M. (2025). "A Deep Learning Framework for Prediction and Inverse Design of Nanoscale Optical Forces near Gradient Metasurfaces." 2025 Nineteenth International Congress on Artificial Materials for Novel Wave Phenomena (Metamaterials), IEEE, X-389.
• Zahraii, P.A. (2025). "Implementation of Residual Tandem Neural Networks for Photonic Inverse Design." Master's Thesis, Chapman University.
• Zahraii, P., Kashanchi, S., Estakhri, N.M., Estakhri, N.M. (2025). "Machine learning-assisted characterization of optical forces near gradient metasurfaces." SMT 2025, APS.

Education

M.S., Electrical Engineering and Computer Science - Chapman University

May 2025 | GPA 4.00

• Thesis: Implementation of Residual Tandem Neural Networks in Inverse Photonic Design
• Valedictorian; Graduation Keynote Speaker

B.S., Computer Science (Neuroscience Minor) - Chapman University

May 2024 | GPA 3.89

• Graduation Keynote Speaker