Austin Mcever, Phd Email & Phone Number

California, United States

São Paulo, Sp, Br

Palo Alto, California, Us

- Developed label collections strategy for optimal training, validation, and test sets for BERT-based transaction categorizer with blend of rules, search, and human based labels- Increased exact match accuracy of categorizer by 15% using data blending, hyperparameter searching, and architecture improvements- Tuned and improved GPT style LLMs for a variety of tasks including Customer Relation Management, persona assignment, and credit default prediction- Lead credit default prediction proof of concept (POC) project with Nubank, pretraining and finetuning large language credit models (LLMs), ultimately resulting in Hyperplane's acquisition

Menlo Park, Ca, Us

Santa Barbara, Ca, Us

Dissertation titled "Detection and Segmentation Using Less Supervision"Learn more about my research on my personal website: www.austinmcever.comHighlights- Collaborate with climate change and marine science group (MARE), to use convolutional neural network (CNNs) models for automation of annotation of videos via context-driven tracking of invertebrate species in natural ecosystem with 14% increase in detection mAP - Released videos as open-source big data set with analytics after additional data collection

- Enhanced Facebook video recommendation modeling in distributed computing environment, driving performance (e.g. watch time) by 1% via A/B testing experiments resulting in full time job offer following graduation- Committed >1,000 lines of code to improve Facebook Watch and Facebook Video Chaining recommendations by increasing performance on low traffic video with data driven approach

Santa Barbara, California, Us

- Adapted YOLOv3 codebase to search for classes of marine vessels in xView dataset of satellite imagery resulting in hundreds of thousands of dollars of funding from Office of Naval Research (ONR) for further applied research

San Leandro, Ca, Us

- Connected OSISoft's PI System data infrastructure to Esri ArcGIS to facilitate a real time data visualization of CURENT's real time power grid simulation- Compared the OSISoft PI System and Esri ArcGIS visualization with custom Python visualization software developed during time with CURENT

Knoxville, Tn, Us

- Created custom visualizations for power system simulations using wxPython- Traveled to Southeast University in Nanjing, China to assist in developing a genetic algorithm to find solutions to microgrid power system simulation stability issues

Doctor Of Philosophy - Phd, Computer Science

Master Of Science - Ms, Computer Science

Bachelor Of Science - Bs With Honors, Computer Science