Resume

Sign in

Machine Learning Engineer

Location:
Portland, Oregon, United States
Salary:
Negotiable
Posted:
March 27, 2019

Contact this candidate

Resume:

Education

PORTLAND STATE UNIVERSITY, BACHELOR OF COMPUTER SCIENCE — GRADUATED IN MARCH 2019

- Focus on Machine Learning, Mathematics, and Programming

- Relevant courses include: Advanced Topics in Machine Learning, Applied Linear Algebra, Applied Differential Equations, Calculus 1-4, Applied Group Theory

- Major GPA: 3.79

Skills

Highly proficient in Python, Julia, C++, and various machine learning frameworks including TensorFlow, PyTorch, and Flux.

Projects

POLICY GRADIENTS, PERSONAL — 2019

LANGUAGES AND FRAMEWORKS USED: PYTHON, JULIA, C++, PYTORCH, TENSORFLOW, FLUX, OPEN AI GYM Implemented the policy gradient reinforcement learning algorithm to solve a wide variety of environments in the Open AI Gym library. The goal of this project was to compare different machine learning libraries across several metrics. To achieve this the project was rebuilt in Python using both PyTorch and TensorFlow, Julia using Flux, and C++ using PyTorch.

SENIOR CAPSTONE, NIKE — 2019

LANGUAGES AND FRAMEWORKS USED: PYTHON, TENSORFLOW, PANDAS, DJANGO Utilized deep convolutional neural networks and transfer learning to train a model on the deep fashion dataset that can recognize the category, color, and attributes of clothing in images. Taught machine learning to several students in my group, helping them to build high performance models both in terms of accuracy and training time. HUMAN PROTEIN ATLAS IMAGE CLASSIFICATION, KAGGLE — 2018 LANGUAGES AND FRAMEWORKS USED: PYTHON, TENSORFLOW, PANDAS, MATPLOTLIB Built and compared several deep convolutional neural network architectures to classify mixed patterns of proteins in microscope images. Wrote significant report with collaborators detailing our findings. MELANOMA CLASSIFIER, PERSONAL — 2018

LANGUAGES AND FRAMEWORKS USED: PYTHON, JAVASCRIPT, TENSORFLOW, DJANGO, REACT Created a drag and drop user interface for designing neural network architectures. It allowed for training models on distributed machines with GPU acceleration from the browser. The results of the various architectures could be tracked as a function of training time and number of epochs to quickly iterate on the best models. Experience

SOFTWARE DEVELOPER, WFG NATIONAL TITLE INSURANCE COMPANY — APRIL 2016 - SEPTEMBER 2017 LANGUAGES AND FRAMEWORKS USED: PHP, JAVASCRIPT, WORDPRESS, REACT, NODE Became sole developer after two weeks of training. Quickly built understanding of the large codebase and maintained several company websites. Initiated and implemented several automation pipelines. Created large time savings by removing the need for manual updates as changes occurred throughout the company, instead these automatically propagated to the correct parts of each website.

Udemy Certificates

- Advanced AI: Deep Reinforcement Learning in Python

- Deep Learning: Advanced Computer Vision

- Deep Learning: Advanced NLP and RNNs

- Deep Learning A-Z: Hands-On Artificial Neural Networks References

CTO, WFG NATIONAL TITLE INSURANCE COMPANY — GORKEM KUTERDEM ac8wyy@r.postjobfree.com

Adam 971-***-**** Kowalski Portland OR ac8wyy@r.postjobfree.com https://github.com/adam-r-kowalski https://linkedin.com/in/adam-r-kowalski



Contact this candidate