Reza Hashemi
San Diego,CA
Summary
Accomplished Lead Scientist at Fair Isaac Corporation, specializing in machine learning and deep learning. Achieved a 120x performance improvement in fraud detection algorithms using PySpark on AWS SageMaker. Expert in cross-functional collaboration and model optimization, driving impactful solutions that meet stringent regulatory standards.
Overview
9
9
years of professional experience
Work History
Lead Scientist
Fair Isaac Corporation (FICO)
San Diego, CA
08.2022 - Current
- Implemented parallel processing for fraud detection algorithms using PySpark on AWS SageMaker, analyzing 800 million records in approximately 40 minutes-achieving 120x performance improvement
- Developed interpretable latent feature-based neural network with PyTorch for non-linear feature extraction, enhancing model explainability for regulatory compliance
- Performed hyperparameter optimization using Optuna and Ray Tune, improving model accuracy by 15%
- Applied protocol buffers in Java to serialize complex data structures efficiently for production systems
- Collaborated with cross-functional teams including product managers, engineers, and compliance officers to deploy ML solutions meeting strict business and regulatory requirements
- Built end-to-end ML pipelines from data ingestion through model deployment and monitoring
Senior Research Scientist
Intelligent Automation Inc.
Rockville, MD
06.2019 - 08.2022
- Developed regression models using ensemble methods to predict mobile traffic patterns in LTE networks with 92% accuracy
- Applied advanced signal processing and statistical methods for noise estimation and pattern recognition in RF data
- Created interactive geospatial visualizations and dashboards for complex data analysis using Python libraries
- Performed exploratory data analysis and feature engineering on large telecommunications datasets
- Led technical contributions to two successful SBIR proposals for the Department of Defense, securing $1.5M in research funding
- Conducted statistical hypothesis testing and A/B testing for system performance validation
Senior Engineer
Automated Precision Inc.
Rockville, MD
08.2016 - 06.2019
- Automated defect detection and classification on industrial parts using convolutional neural networks (CNN), achieving 95% classification accuracy
- Developed computer vision algorithms for real-time object detection and quality control systems
- Implemented GPU-accelerated computing using CUDA for processing point cloud data from LiDAR systems, reducing processing time by 80%
- Built data preprocessing pipelines for cleaning, normalizing, and augmenting image datasets
- Applied dimensionality reduction techniques (PCA, t-SNE) for feature analysis and visualization
- Performed A/B testing and statistical validation of ML models in production environments
Education
PhD - Electrical Engineering
University of Arkansas
FayettevilleSkills
- Machine Learning
- Deep Learning
- Neural Networks
- PyTorch
- TensorFlow
- Statistical Analysis
- Modeling
- High-Performance Computing
- Parallel Processing
- PySpark
- CUDA
- Cloud Computing
- AWS SageMaker
- Computer Vision
- Pattern Recognition
- Feature Engineering
- Model Optimization
- Explainability
- Big Data Analytics
- Python
- Java
- C
- Cross-functional Collaboration
- SQL
- Git
- SVN
- Scikit-learn
- Keras
- XGBoost
- LightGBM
- AWS EC2
- AWS S3
- Distributed Computing
- Pandas
- NumPy
- OpenCV
- Matplotlib
- Seaborn
- Plotly
- Optuna
- Ray Tune
- Hyperopt
- Grid Search
- Bayesian Optimization
- Deep Neural Networks
- CNNs
- Ensemble Methods
- Regression
- Classification
- Clustering
- Dimensionality Reduction
- Time-Series Analysis
- Parallel Computing
- A/B Testing
- Model Explainability
- SHAP
- LIME
Awards Recognition
- ASEE/NSF Small Business Postdoctoral Research Diversity Fellowship
- Honorable Mention Paper, IEEE Intl. Symposium on Antennas and Propagation
- John A. White Faculty-Student Collaboration Award
- NSF Award, San Diego Supercomputer Center
- Listed in Marquis Who's Who in America and Who's Who in Science and Engineering
Projects Achievements
- Developed production-grade ML models processing 800M+ records with sub-hour latency
- Built interpretable neural networks for regulated industries requiring model explainability
- Achieved 95% accuracy in automated defect detection using computer vision
- Secured $1.5M in research funding through technical leadership and proposal writing
- Published 15+ peer-reviewed papers in machine learning and computational methods
Selected Publications
- Non-invasive detection of optical changes elicited by seizure activity using time-series analysis, Journal of Neuroscience Methods, 227, 2014, 18-28
- Noninvasive evaluation of nuclear morphometry in breast lesions using multispectral diffuse optical tomography, PLOS ONE, 7, 9, 2012, e45714
- High performance computing for the level-set reconstruction algorithm, Journal of Parallel and Distributed Computing, 70, 6, 2010, 671-679
- Shape Reconstruction Using the Level Set Method for Microwave Applications, IEEE Antennas and Wireless Propagation Letters, 7, 2008, 92-96
Personal Information
Citizenship: U.S. Citizen
Timeline
Lead Scientist
Fair Isaac Corporation (FICO)
08.2022 - Current
Senior Research Scientist
Intelligent Automation Inc.
06.2019 - 08.2022
Senior Engineer
Automated Precision Inc.
08.2016 - 06.2019
PhD - Electrical Engineering
University of Arkansas