Ann Arbor, MI. Personalization at scale.

Parker Moesta.

Senior Machine Learning Scientist

Shipping end to end recommender systems.

See my work Resume

Selected Projects

My Work

I like building things end to end. Custom deep learning architectures, recommender systems, computer vision, NLP, network analytics. Every card links to the source.

Milestones

How I Got Here

Economics, then enterprise tech, then production ML. The short version.

  1. 2016

    B.A. Economics, University of Michigan

    Graduated and headed to Boston to figure out what I actually wanted to do.

  2. 2016 to 2018

    Oracle, Customer Facing Role

    Learned the business ropes. Worked directly with enterprise customers and led a chatbot proof of concept for a retail partner, my first taste of shipping something intelligent.

  3. 2018 to 2020

    The Retooling

    Went back to fundamentals. Python, statistics, and the math underneath. Coauthored simulation research published in IEOM 2021.

  4. 2020 to 2023

    Data Scientist, Coupa Software

    Built my first production recommender system and prototyped an early question answering assistant. Learned what it takes to keep a model alive in production.

  5. 2023

    Domino's, Digital Experience Team

    Causal inference and A/B testing across web and mobile. Shipped a deep learning model for cart abandonment.

  6. 2024

    M.S. Data Science, University of Michigan

    Finished the degree while working full time. Generative models, NLP, and computer vision. Most of the projects above were born here.

  7. 2025

    Personalization Team

    Helped build the feature store and the company's first production recommender systems.

  8. 2026

    Senior Machine Learning Scientist

    Recommenders end to end, from research idea to production system. Still having fun.

Parker Moesta outdoors on a wooded trail

About

Business problem first.
Research second. Production always.

I am a Senior Machine Learning Scientist at Domino's. I lead applied ML and production architecture for eCommerce personalization, the systems that decide in real time what millions of customers see. My work spans distributed feature engineering over billions of records, two stage recommender systems, transformer sequence models, and the experimentation and causal inference to understand what actually works.

Detroit born, Ann Arbor based. I came to ML through economics and enterprise tech, and that path shaped how I operate. When I am not shipping models, I am in the mountains, or teaching a neural net to write recipes.

Now
Senior ML Scientist, Domino's
Focus
RecSys, Experimentation, Production ML
Education
M.S. Data Science and B.A. Economics, University of Michigan
Previously
Coupa Software, Oracle
Published
IEOM 2021, featured in a Databricks technical blog
Download resume

Contact

Let's talk.

Email is fastest. I read everything.

parkermoesta@gmail.com
Neural Collaborative Filtering framework

Steam Game Recommender, Neural Collaborative Filtering

A personalized game recommendation system for Steam users built on Neural Collaborative Filtering, trained on interaction data with playtime as an implicit preference signal.

Live demo GitHub
Neural style transfer result

Neural Style Transfer

A Neural Style Transfer model built on the VGG19 architecture, trained to apply the style of one image to another while preserving the original content.

GitHub
Faces generated by the VAE

Novel Face Generation, Variational Autoencoder

A VAE trained on CelebA for 50 epochs that generates novel faces, visualizes the latent space distribution, and manipulates generated images through latent arithmetic.

GitHub
WGAN-GP architecture

Conditional WGAN-GP for Animal Face Generation

A Conditional Wasserstein GAN with gradient penalty trained on a dataset of animal faces, generating novel faces conditioned on class.

GitHub
Grid of colorized historical images

pix2pix, Restoring Color to Historical Archives

Image to image translation with a UNet generator and PatchGAN discriminator, trained on paired historical and modern photos to colorize grayscale images of houses around Ann Arbor while preserving texture and detail.

GitHub
LSTM architecture

Novel Recipe Generator, LSTM

An LSTM network trained on a large recipe corpus that generates novel, creative recipes. A hands on study in sequence generation.

Try the model GitHub
News word cloud

BERT Sentiment Analysis of the News

A sentiment analysis pipeline for climate coverage. NewsAPI for retrieval, NLTK for preprocessing, a pretrained BERT model for sentiment, and visualizations that surface trends across outlets.

Read on Medium GitHub
BERT classifier diagram

News Source Classifier, Fine Tuned BERT

A classifier that predicts an article's outlet from its text using a pretrained BERT backbone. It reached 95 percent accuracy after 10 epochs on CNN, Fox News, MSNBC, and Breitbart.

Read on Medium GitHub
t-SNE embedding projection

Word2Vec and t-SNE, Semantic Geometry of the News

Trains Word2Vec embeddings on articles from different outlets and projects them with t-SNE to visually compare how sources encode meaning around the same topics.

Read on Medium GitHub
Animated growth of the PyTorch citation network over time

PyTorch Research Network Dynamics

Network analysis mapping the PyTorch research landscape. Coauthorship communities, key figures, and a multilevel citation network tracing how the original PyTorch paper propagated through the field.

GitHub
CO₂ change rate chart

Climate Finance and the Decoupling of CO₂ from Growth

An econometric investigation of climate related financial flows into developing countries, asking whether these investments helped nations cut emissions while maintaining economic growth.

View the study
Interactive route map through the North Cascades

A Visual Journey Through the North Cascades

A five day backpacking route rebuilt as an interactive map from Apple Health GPS exports and photo metadata. Altitude coded by color, photos pinned where they were taken.

Explore the map GitHub