I've been building software that handles the critical financial operations of today’s political campaigns – fundraising, accounting, and compliance – in one single, simple platform.

I've touched all parts of the stack but have mostly been involved in the backend.

Researched, built out, and maintained an ETL pipeline that processes public giving history records from multi-gigabyte gzipped CSVs in s3 and loads it into an Elasticsearch cluster for extremely fast querying.

Implemented complex filtering on our clients' CRMs by utilizing advanced features of SQL in Postgres and set theory. (ie: find users that (have a phone number AND an address) OR (have donated greater than $100 once last quarter AND have given to any federal campaign last year))

Integrated our product with Sparkpost (email) and Infobip (text) service providers and built a reliable, scalable, and effective communications platform.

Built out frontend features within the CRM product in Typescript and Next.js

Tuned our servers on Heroku.

Migrated billions of rows of text-messaging data from postgres to MongoDB and made it searchable.

Our tech stack includes:

Worked as a Software Engineer on the Voices and Gold Teams. Responsible for backend work on user profiles and the Reddit Gold product.

Maintaining and updating legacy monolith code to extend functionality to automoderator to act on posts and comments made on user profiles. Debugged legacy Pylons and RabbitMQ queue consumer code.

Implemented profile image uploading using pre-signed urls to share objects on s3. Ensured that images uploaded to s3 were scanned with PhotoDNA for security. Communicated to frontend via websocket on successful upload. Images are purged from our CDN on delete, so as not to allow images to linger.

Built out infrastructure to liase with Stripe, Paypal, Apple ITunes Store, and Google Play Store to allow users to purchase Reddit Coins and Reddit Premium on a variety of platforms.

Maintained the Reddit Gold accounting infrastructure. Gathered needs from Accounting team and built out monthly cronjobs to transform poorly formated raw data and generate csvs sent to Accounting via email. Updated Postgres SQLAlchemy models to improve usability of our transaction models.

Advocated for building out tools to improve the quality of life of our internal Community team. Maintained and updated a tool to send out messages en-masse to moderators. Designed and built a tool for digging into a user's coin spending history to better discover fraudulent activity.

Wrote design docs and built out a microservice for Reddit Awards. Server built using Baseplate, deployed with kubernetes/spinnaker, uses Cassandra for data model storage, exposed endpoints via Thrift, implemented queries and mutations on GraphQL to utilize the new service.

Mentored at Hackbright for 2+ years and ended up recruiting one of my mentees as a Software Engineer on my team.

Over the course of 3 years at AdRoll I was promoted to Senior Software Engineer and became the service maintainer of a critical piece of infrastructure. Here's a non-exhaustive list of some of my accomplishments at AdRoll:

Became Service Maintainer of Mirrorman - a system for detecting object model changes and syncing data to various partner networks. Built in SQLAlchemy, using Celery for periodic tasks, it is some of the oldest code in the AdRoll repo and is tightly coupled with a large part of AdRoll code.

Built out AdRoll integration with Shopify and Woocommerce by reverse engineering stores to develop a javascript snippet that capture events on a page. Webserver built in Flask-Sqlalchemy and deployed on Amazon Elastic Beanstalk.

Managed a team dedicated to building an A/B testing framework around our Dynamic Creative ad product recommendations. The project spanned many teams and included modifications to EMR jobs and our HTML5 ad-serving framework.

Built out AdRoll integration with Facebook Website Custom Audiences. Maintained feature parity with Facebook Ads API including mobile app install ads, platform/placement/demographic targeting, performance tracking, etc. Wrote tools in Selenium and PhantomJS to perform batch tasks in a headless browser.

Updated RESTful API endpoints for manipulating the AdRoll object model. Built on Pylons.

Added new pages and fixed old pages of the internal website used by Account Managers. Templates in Mako, Controllers built in Pylons. Some basic HTML, JS, and CSS modifications.

Constantly modified AdRoll object model to suit new business needs. Object model stored in Postgres database, using SQLAlchemy ORM, and Alembic for schema migrations.

Received training to become Scrum Alliance Certified ScrumMaster® (CSM)

Worked in a small team of engineers and designers developing the social discovery site combosaurus.com.

Modified webserver and RPC-servers built on Express (nodejs framework) written in Iced Coffescript (http://maxtaco.github.io/coffee-script/)

Processed Wikipedia, Freebase, and Musicbrainz's large datasets with Python

Integrated site with Goodreads api

Developed email link tracking system with anti-spoofing measures

Helpdesk support for Bondstudio, an software suite of fixed-income product analytic tools.

Built out and maintained a system of data analysis software, designed to scrape and process

gigabytes of data provided by FNMA, GNMA, and FHLMC.

Developed ad-hoc software to provide reports for the MBS research team.