Andrew Borthwick work email
- Valid
- Valid
- Valid
Andrew Borthwick personal email
- Valid
Andrew Borthwick phone numbers
A Principal Scientist with a knack for blending state-of-the-art AI with real-world impact, Andrew has earned a reputation as a leader who doesn’t just push boundaries—he helps reshape entire industries. With over two decades of experience, including founding and scaling a successful ML startup, Andrew has developed groundbreaking solutions in record linkage, supply chain optimization, and data deduplication. Most recently, he spent 10 plus years at Amazon as the lead scientist on projects that brought powerful but easy-to-use "automatic machine learning" techniques to product demand forecasting and large-scale product classification. In addition, he developed a record linkage system for Amazon internal applications ranging from Amazon Prime Video to the Amazon Product Catalog which was later turned into an AWS Service. Andrew is the author of multiple highly cited publications and patents that address the accuracy and scalability of an ML-based record linkage system. His influential doctoral thesis was among the first to apply ML to NLP. At Auger, Andrew aims to leverage his deep expertise in AI and ML to build transformative solutions for supply chain automation, ensuring Auger customers benefit from cutting-edge science that is seamlessly integrated into their workflows.
-
Principal ScientistAuger Dec 2024 - PresentBellevue, Wa, UsI am thrilled to have joined Auger as Principal Scientist! Reinventing the supply chain requires powerful AI and machine learning solutions that are accessible to all, not just scientists. At Auger, I’m excited to integrate this passion with my deep interest in supply chain and product data, creating intuitive solutions that drive smarter, faster decisions. -
Principal ScientistAmazon Web Services (Aws) Apr 2024 - Dec 2024Seattle, Wa, UsAutomatic Time Series Forecasting for Demand Planning• Working in a startup-like environment at the new AWS Supply Chain, I researched time series forecasting methods for product demand prediction.• Identified a highly effective state-of-the-art approach, delivering forecasts more accurate than previous models while requiring less human intervention.• Invented new time series metrics to optimize for demand prediction scenarios, especially for products with erratic demand histories. -
Principal ScientistAmazon Aug 2020 - Apr 2024Seattle, Wa, UsAutomatic MultiModal Classification for the Amazon Product CatalogScientific lead of a group focusing on thousands of classification tasks over the 100M+ items in the Amazon Product Catalog. For instance, "Is this product flammable?", "Does this product claim to cure Covid-19?", "Is this a motorcycle helmet?" The problem space required leveraging signals from image, text, and numeric product metadata to achieve the most accurate decisions.To improve on the prior system, I led a collaboration spanning multiple VP-level teams that introduced a new state-of-the-art (SOTA) AutoML toolkit. We also simplified the system architecture to allow a 3x growth in available training data, leveraged a broader multimodal schema, and conducted careful benchmarking. The result was a 10x increase in automated annotations and corresponding reduction in the need for human intervention. -
Principal ScientistAmazon Web Services (Aws) Apr 2019 - Aug 2020Seattle, Wa, UsLead scientist in the ML Solutions Lab group, mostly focused on building the "FindMatches" general-purpose Record Linkage and Data Deduplication capability in AWS Lake Formation. -
Senior Machine Learning ScientistAmazon Web Services (Aws) Sep 2017 - Mar 2019Seattle, Wa, UsScientific lead of the project that built the general-purpose Record Linkage and Data Deduplication capability known as AWS Lake Formation FindMatches. FindMatches uses machine learning to find and link records that refer to the same entity but don’t share a reliable identifier. -
Senior Machine Learning ScientistAmazon Dec 2014 - Sep 2017Seattle, Wa, Us• Led the effort to build a general-purpose system for record matching and deduplication across multiple domains, including products, businesses, addresses, and video content• Gave a prestigious internal "Principals of Amazon" talk about this project, which was highly rated and seen by hundreds of people -
Principal Software Development Engineer (Bing Local)Microsoft Feb 2013 - Nov 2014Redmond, Washington, UsSynthesize information from information providers and web crawls into a single global index of local businesses and professionals that is accurate and free of duplicates -
Principal Scientist And Director Of Data ResearchIntelius 2009 - Dec 2012Seattle, Wa, Us• Create and hire the Intelius' Data Research Group, a team of top scientists in machine learning and computational linguistics focused on record linkage, information extraction, and people search• Research, develop, and deploy a general pipeline to process all Intelius people data and to link any Intelius profile to any other Intelius profile• Research, develop, and deploy processes to extract biographical information about people from a general crawl of the web to generate profiles for Intelius websites -
Independent Expert WitnessBoies, Schiller & Flexner Llp Aug 2008 - Mar 2009New York, Ny, UsHoffman vs. American Express: Expert witness on people matching defending AmEx in $310M class action suit. Ran multiple experiments on AmEx data, prepared 16-page report and gave a deposition on a portion of the case involving alleged charges for uninsured passengers. Judge ruled in favor of AmEx on all counts after 11-week presentation by plaintiffs. -
Principal ScientistSpock Networks 2007 - 2009Us• Developed technology which was critical to Spock's acquisition by Intelius, Inc.• Research and develop processes to extract biographical information about people from diverse public websites to generate profiles for the Spock website• Research and develop process to link profiles of people so that the Spock website will have only one profile for each real-world individual• Research and develop person name spelling correction algorithms to suggest alternatives when a person name is misspelled -
CeoChoicemaker Technologies 1998 - 2006Business Accomplishments• Founded the company to commercialize a machine learning approach to the problem of approximate record matching and database deduplication.• Coordinated successful deployments for customers including the Centers for Disease Control, New York City Department of Health, and nine state education departments• Grew the company to 12 peopleTechnical Accomplishments• Principal inventor of multiple patents for using ML in record matching and ensuring system scalability• Principal investigator of three NSF Small Business Innovation Research grants totaling $1M. -
Systems ConsultantMorgan Stanley 1993 - 2002New York, Ny, UsWorking only one day per week, was critical designer and maintainer of Information Services Allocation Model (ISAM), which solves matrix algebra equations to equitably allocate over $1 billion in IT costs to the rest of the firm. -
Programmer/Team Leader/Business AnalystMorgan Stanley 1988 - 1993New York, Ny, Us* Designed and managed the project which built ISAM* Supervised a team of four to take the project from a few pages of notes to a finished project* Personally coded the mathematical heart of the system in APL
Andrew Borthwick Skills
Andrew Borthwick Education Details
-
New York UniversityComputer Science -
Oberlin CollegeHistory
Frequently Asked Questions about Andrew Borthwick
What company does Andrew Borthwick work for?
Andrew Borthwick works for Auger
What is Andrew Borthwick's role at the current company?
Andrew Borthwick's current role is Principal Scientist.
What is Andrew Borthwick's email address?
Andrew Borthwick's email address is ae****@****ail.com
What is Andrew Borthwick's direct phone number?
Andrew Borthwick's direct phone number is +165031*****
What schools did Andrew Borthwick attend?
Andrew Borthwick attended New York University, Oberlin College.
What are some of Andrew Borthwick's interests?
Andrew Borthwick has interest in Emc Corporation, Argo (2012 Movie), Blue C Sushi, Facebook Seattle, Oberlin College.
What skills is Andrew Borthwick known for?
Andrew Borthwick has skills like Machine Learning, Natural Language Processing, Python, C++, Algorithms, Java, Data Quality, Record Linkage, Computational Linguistics, Agile Methodologies, Distributed Systems, Databases.
Free Chrome Extension
Find emails, phones & company data instantly
Aero Online
Your AI prospecting assistant
Select data to include:
0 records × $0.02 per record
Download 750 million emails and 100 million phone numbers
Access emails and phone numbers of over 750 million business users. Instantly download verified profiles using 20+ filters, including location, job title, company, function, and industry.
Start your free trial