Michael Mastroianni's Email Addresses & Phone Numbers

📖 Summary

Machine Learning Engineer @ Mercari Mostly working on personalization: • Generating preferences for brands and cagegories from log data • Running alternating least squares on the results • Upload results to GCS for use by front end recommender services • Generating search similarities using word embeddings and creating user profiles in the same space for doing search recommendations Cambridge, MassachusettsContract Data Scientist/Architect @ Self Employed For a SAS company doing chatbots: • Did NLP and Predictive Modeling to analyze chats • Developed and productionized models for chat classification into sales/support/spam • Wrote visualization code for viewing distribution of chats in classes • Developed and productionized automatic topic inference for chats using LDA and NNMF • Developed and productionized sentiment analysis module • Wrote visualization code so that a customer could easily see the most common chat topics and sentiments • Wrote ElasticSearch ingest plugins for named entity resolution and topic analysis (recognizing companies, people names, and locations) • Wrote ElasticSearch analyzer plugins for dealing with the fact that chats are formatted and typed differently than normal textFor a company doing a blockchain/p2p search system in China: • Developed architecture for peer to peer network service • Set up automated builds where there had been none • Converted prototype code from maven to gradle • Found mirrors for mavencentral in China so that builds could happen in a reasonable timeframe • Wrote concurrency, peer discovery, peer updating, and queuing code for peers • Developed architecture of the blockchain subsystem • Productionized python models for document classification for real time querying via REST using flask, gunicorn, and docker From January 2018 to September 2018 (9 months) Boston and BeijingArchitect/Data Scientist @ Placester, Inc. • Overall data design for search and segmentation. • Development of natural language search for real estate from end to end from scratch. • Hadoop based segmentation based on search logs. • Integration of ads into search. • Some geo-processing for real estate listings. From June 2013 to February 2017 (3 years 9 months) Boston, MAData Scientist/Architect @ PayPal Where was acquired by PayPal.• Wrote pipeline of hadoop jobs for turning transaction logs into category preferences for generating recommendations.• Primary author of open source LSH used to generate recommendations from this data: see http://code.google.com/p/fast-lsh/• Wrote churn prediction system using a series of hadoop jobs and streaming classifiers.• Wrote back end of ad targeting system, allowing business users to generate ad campaigns based on sets of exemplar merchants or customers.• Team lead for new customer acquisition project based on mining third party data. From May 2011 to June 2013 (2 years 2 months) Boston, MAArchitect - Big data and Recommendations @ WHERE, LLC Where built a location aware advertising platform for smartphone apps, as well as a smartphone app for android and iPhone which is still available. Where was acquired by eBay/Paypal in May of 2011.• Wrote search portion of recommendation system using Hadoop for data processing/transformation and Lucene as the basic search engine.• Wrote kd-tree based lookup system for finding nearby places• Wrote hadoop/mahout-based item similarity recommendation system using user/location preferences as expressed by ratings.• Incorporated this recommender into a lucene based search engine, with location as part of the preference calculation• Wrote system based on LSH for deduplicating place data. From August 2010 to May 2011 (10 months) Boston, MASr Software Engineer @ Gerson Lehrman Group • Wrote much of search implementation in java using lucene.• Wrote a system for switching/updating indices.• Wrote server skeleton, with basic thread management, pluggable services, and xml configuration in java, using jetty as an internal http server.• Wrote Thread management system used java.util.concurrent, and extended it to ensure that all threads/threadpools had registered shutdown hooks.• Designed and wrote the More Like This feature for finding related search results. From March 2006 to July 2010 (4 years 5 months) Boston, MASr Software Engineer @ ChoiceStream • Wrote most of the personalized websearch product. This project involved doing searches, spidering web pages, scoring them wrt a set of attributes, using these scores to score against a user profile, and finally presenting the pages in an order determined by this score.• Wrote the threadpool and basic synchronization code used throughout the codebase• Designed and built the production attributization engine (which generated sets of attributes for web pages) and the profile-based recommender of the websearch product.• Wrote a tokenizer and clustering engine for the sponsored links product. From 2003 to 2005 (2 years) Cambridge, MASr Software Engineer @ Predictive Networks • Algorithm development and implementation for SmartSurf, which used collaborative filtering to suggest pay-per-view titles to cable TV users.• wrote, maintained and extended software that served targeted advertisements to client software on set-top boxes. • Wrote significant portions of software that served targeted advertisements to users of the Predictive client. • Wrote much of the Predictive Client, used by the AT&T WorldNet service to serve personalized ads to internet users. The client was written in C++. It logged web traffic, communicated summaries of this traffic to a server, and displayed targeted advertisements sent down by the server.• Wrote auto-update module for this client, which successfully upgraded 3.5 million AT&T customers From 1999 to 2003 (4 years) Cambridge, MASoftware Engineer @ Biztravel • Wrote a neural network based system to predict credit card fraud• Wrote a module for weather tracking• Wrote code for caching of airport, flight, and hotel information• Wrote code for email notification of ticket purchases, registrations, and session tracking• Wrote code for finding and storing data for mileage accounts, traffic logging• Wrote several spiders/scrapers From 1997 to 1999 (2 years) Cambridge/Somerville, MASenior Architect @ CarGurus • Used word2vec embeddings to detect synonyms for search term bidding, in order to increase the number of terms we bid on • Used models trained on past ad performance to bid on items we hadn’t seen before, rather than using heuristics • Set up and maintained production python environment • Developed and productionized time series analysis for bid adjustments in python using ARIMA • Productionized models built in R and Python • Developed and productionized LSTM model for anomaly detection for various ads metrics, so that we could get automated alerts when problems arose • Bug fixes and minor features for the main Ruby codebase From February 2017 to December 2017 (11 months) Cambridge, MA