I am the manager for the Access & Collection team

I work on the Access and Collection team, focusing on integrating foreign data sources into our platform.

I've transitioned to the machine learning projects on Places, participating in feature engineering, model training, and model deployment into production Places assets.For the first year, I've have the pleasure of helping with integration efforts from Foursquare's Places asset and Factual's Places asset. There have been many roadmaps, proposals, brainstorms, "can I sync with you?"'s on completely foreign tech stacks and idiosyncrasies along the way. I play an active role in the coordination of various integration efforts and steps, while continuing to drive roadmapping and goal-setting for ingestion and chains-related processes.

"Founded in 2015 by Sebastian Wolff, Materia Collective began as a collection of musicians seeking to celebrate the music of Final Fantasy with a massive arrange album. Materia Collective has since grown into a label that produces and releases original music alongside creative interpretations and arrangements of new and classic game scores.Today, Materia Collective features an international array of musicians, brought together by a passion for gaming, music, and creative entrepreneurship to continue developing the unique identity of the group."https://www.materiacollective.com/artist/ann_zengContributions to 8 albums and counting.

As a member of the Places team and technical lead to ingestion efforts, I help to monitor and improve the quality of our Global Places dataset as it travels through our data processing pipelines. I help create and maintain web crawls, organize and ingest our system of chains and chains-relevant POI data, and trusted partner contributions. All of these millions of inputs are shepherded (by myself and other data engineers) through internal systems until they become part of our global Places datasets. When there are data regressions, I dig through the data through whatever means possible (sometimes it's a bunch of API calls, sometimes it's flipping through other reports and merge histories, and sometimes it's knowing the right people to talk to) to design the proper solutions or set the proper expectations for Product and our customers. On any given day, I might be reading or writing Ruby, Python, Scala, or Apache Spark flavors of languages. Pull requests have been made to internal tools, to big data processing workflows, and dashboard or report-generation scripts (there's of course also the occasional one-liner or two that everyone complains about but nobody prioritized enough to just do it).I spent two quarters as tech lead and individual contributor (plus a third hat, informal technical product manager alongside our actual product managers!) roadmapping goals and initiatives for all things Places ingestion, helping our team achieve our company-reported objectives and key results during those two quarters. I am also the main IC/lead for chains, a subproject within ingestion with high impact across all of our other products.

As a PHP developer on the Feed Development team, I support the Feedonomics Fulfillment team by writing web-scraping scripts that help our clients better integrate with the Feedonomics platform. I also generate and process data feeds for our clients, as well as do data cleanup for them. I'll work with the APIs of various Ecommerce platforms, like Bigcommerce or Shopify, or shopping engines and shopping marketplaces, like Google, Amazon, etc. Sometimes, I write SQL queries to show KPIs to management or to help further organize client data.

At Feedonomics, we present a full-service solution for your product feed needs. Every potential client gets to speak with the Technology Integration team, as we give feedback on how the Feedonomics platform can integrate with their current Ecommerce solutions. I consulted with these potential clients on integrating with Feedonomics as a sales engineer/support role. As a part of this process, I compiled and presented audits to both potential clients and signed clients regarding the quality of their current shopping feed, or I put together tailored technical demonstrations using client-given data. I also coordinated developer requests and resources tasked out by members of the fulfillment team. Additionally, I oversaw or supported many enterprise/key accounts of Feedonomics, providing technical support and coordinating developer requests, as these clients use our platform for their data feed needs. To assist them, I also helped them clean up and organize their data using the Feedonomics platform. In my spare time, I also pioneered projects like new feed analyst training documentation and self-serve client training documentation.

Publication: Michael, E. B., Bradley, P., Hefright, B., Saner, L., Mishler, A., Zeng, A., & Danks, J. H. (2018). Human use of machine translation to extract information from texts. In I. Lacruz & R. Jääskeläinen (Eds.), Innovation and expansion in translation process research. Amsterdam: John Benjamins.During my summer at CASL, I worked as a research assistant on a project which explores the interactions of machine translation and human translation (so, Google versus a human brain and a dictionary). With the design of the experiment already in place, the team and I began work running and evaluating the current study. I recruited language learners from across the United States and identified a novel method of quantitatively evaluating our participants' results, based off of previous similar studies. I also provided additional research, reports, and publication reviews on Translation Memory, a type of translation tool.Aside from my assigned research project, I participated in a weekly python seminar taught by one of the research scientists at CASL. Summoning up my first-year computer science courses, I helped provide a small additional seminar on the basics of object oriented programming.

During my senior year in college, I participated in a year-long internship with members of the Fulbright Program at the US Embassy in Beijing. Despite our 16-hour time difference and 6000+ mile location difference, we put together a number of different projects throughout the 10 months together.The first project was to edit and proofread the directory of that year's Fulbright participants. My bilingualism came in handy, as I could read both the Chinese and English portions of the directory. The second project was to create and manage the marketing materials for Fulbright conferences and other events. An old brochure was revamped with new images, testimonials, and text to be handed out throughout the year at all the events that the Fulbright staff would attend and host. The last project involved QAing and layout tweaks to the current Fulbright website. Different programs had their various content pages revamped with updated data and standardized with respect to each other to ensure a better user experience scrolling through the site.

Once for Syntax I (Winter quarter 2014), once for Syntax II (Spring quarter 2015).Since syntax is my favorite linguistic subject, what better way to fully understand it than to teach it? For two separate teachers, I helped put together lecture materials and homework assignments for the students. In my own time, I put together additional guides and problem sets for midterm and final exam review. I held office hours on my own, and put together additional review sessions for the midterm and the final, where we went over homework problems, lecture concepts, and novel problem sets as well.

Led by RedHen Labs and Communication professor Frances Steen, this project aims to aid in the development of algorithms used to help make tagging of closed captions more accurate and efficient.I was brought onboard as one of the first research assistants, and I demonstrated enough Unix knowledge at our initial interview that my first task was to create a manual for the rest of the RAs who would join our team. The manual taught them how to install Python, get used to working with a Unix shell, and use the basic Python script that would help them perform their research duties. Once the team was assembled and the software was acquired, we all worked to hand-tag closed captioned transcripts using the Penn Treebank tagging guidelines. These transcripts tagged with parts of speech were to be used as learning data for the machine learning algorithm that was being developed by the lab.

Linguistics research involving African tonal languages: Organize written fieldwork data of the Seeku language collected by graduate student.Cut and organize audio files of fieldwork data of the Seeku language collected by graduate student.Transcribe folk music of the Tommo So language collected by graduate student.

Hitachi Data Systems' Global Solution Services aims to help enterprise customers in protecting and managing their data. During my summer there, my main project was to revamp and update their internal website resources used by the field engineers. Their current website system was more like a forum-style system with disorganized pages, so I took inventory of the existing pages and began to create both standardized landing pages and content pages that would consolidate the scattered information. Team members both in the office and in the UK worked with me to put together the new webpages. Their service manuals, solutions documents, process checklists, etc. were updated and rolled out during that same summer for some of the most commonly provided services.An additional project I picked up that summer was upkeep of the HDS community site for partners, which mirrors internal website resources that were nonconfidential.