Aaron Beyer
Professional Services Data Engineer at Amenity Analytics- Claim this Profile
Click to upgrade to our gold package
for the full feature experience.
-
English Native or bilingual proficiency
Topline Score
Bio
Credentials
-
AWS Certified Solutions Architect – Associate
Amazon Web Services (AWS)Jul, 2023- Nov, 2024 -
Excel 2016: Advanced Formulas and Functions
Lynda.comJun, 2016- Nov, 2024 -
Excel 2016: Advanced Formulas and Functions
Lynda.comJun, 2016- Nov, 2024 -
Excel 2016 Essential Training
Lynda.comMay, 2016- Nov, 2024 -
Data Analysis
Dataquest.io -
Data Visualization
Dataquest.io -
Machine Learning Subject
Dataquest.io -
Machine Learning Subject
Dataquest.io -
Statistics
Dataquest.io
Experience
-
Amenity Analytics (part of Symphony)
-
United States
-
Software Development
-
1 - 100 Employee
-
Professional Services Data Engineer
-
Mar 2019 - Present
--Enhanced the efficiency and reliability of data distribution by building, maintaining, deploying, and monitoring data pipelines that obtain, modify, and distribute data using Python, AWS CDK, Lambda, S3, RDS, ECS, Docker, and CircleCI which ensures consistent and timely data availability, augmented by comprehensive testing (unit/integration/e2e)--Created a robust and centralized platform for sending daily and historical backtesting and ESG files to clients by designing and implementing a service leveraging AWS DynamoDB, Athena, S3, Lambda, CloudWatch, SQS queues, Step Functions, and Google Big Query, which lowered the time it takes to add a new client and create historical files from days to minutes --Expanded and maintained API gateway endpoints and assisted in the deployment of the service on customer servers which facilitated adoption of the company’s services--Improved service speed, cost-efficiency, and reduced errors in our main lambda caching mechanism by revamping the caching methodology to align with cloud-native principles which achieved a 99% reduction in errors, 90% increase in speed, and 80% reduction in operational costs--Minimized potential customer churn due to data inconsistencies by conducting deep data analyses to pinpoint and rectify integrity issues--Created a service that allowed testing of changing NLP components before changes are pushed to production, saving model developers 20-30 hours a month Show less
-
-
Data Scientist
-
Feb 2018 - Mar 2019
--Wrote rules to extract important information from news and financial documents--Identified language patterns to extract key insights from news and financial documents--Trained language models like BERT on NLP extractions from financial documents to determine if models were extracting the right information--Analyzed extractions from financial documents for discrepancies and insight using Python libraries like Spacy, NLTK, and Gesnim--Designed and implemented topic modeling analyses using word2vec and TF-IDF Show less
-
-
-
Metis
-
United States
-
Higher Education
-
1 - 100 Employee
-
Data Science Student
-
Apr 2017 - Jun 2017
Metis is a 12 week, immersive data science bootcamp covering topics such as machine learning, statistics, programming, communication and visualization. Obtained data and created projects involving supervised and unsupervised learning, NLP, classification, and regression. Projects: Beating the Club: Tuned a random forest, a logistic regression, and other classification models with grid search to predict who will pay off their loan at LendingClub.com. Brought the precision score of predicting a good loan to .92. Created an easy to use Flask app to help decide which loans someone should invest in. We Need to Talk about Mental Health: Used Reddit's API to obtain posts and comments from mental health Subreddits and stored them on MongoDB. Then used topic modeling and sentiment analysis to analyze the ways Redditors talk about mental health. Also further analyzed posts to try to determine mental health issue severity. Takeaways: comments tend to be more positive than posts, posts mainly centered around telling traumatic stories or asking for help, and more positive posts tend to be more specific in discussing types of treatment than more negative posts. The Anatomy of an Outrage: Used Twitter's API to get Tweets in reference to a divisive news story. Partitioned Tweets based on time and analyzed the progression of the story using topic modeling, sentiment analysis, and political leaning analysis. Interesting findings: there were a significant amount of bots (around 40% of the tweet traffic) , many seemed to lean conservative (about 90% of the bot tweeters), and the bot traffic probably was a large cause of why the story went viral. The Gross Decline: Scraped BoxOfficeMojo and used the OMDB API to obtain film information. Cleaned erroneous rows, and created new features to enhance prediction. Then used linear regression with elastic net regularization to predict the decline in box office gross between week 1 and week 2. Show less
-
-
-
Element Fleet Management
-
Canada
-
Financial Services
-
700 & Above Employee
-
Data Specialist | Customer Lead
-
Aug 2016 - Mar 2017
--Worked directly with clients to locate vehicles and help cut spending by optimizing maintenance and transportation costs. --Cleaned, manipulated and organized data so fleet managers could have a better understanding of vehicle alignments and fleet costs --Used advanced excel tools like pivot tables, VBA, VLOOKUPs and conditional formatting to verify and fix incorrect data during data migration process --Assisted fleets in dealing with driver and vehicle issues including maintenance troubles, rental car needs, fuel card issues, and licensing issues --Trained and mentored new team members in using fleet systems and in fleet terminology Show less
-
-
-
Target
-
United States
-
Retail
-
700 & Above Employee
-
Fraud Analyst
-
Nov 2015 - Feb 2016
--Handled inbound and outbound calls in order to help guests with card issues and to deter fraudsters --Analyzed transaction histories and account information for possible credit card fraud --Used credit bureau software and public records to verify sensitive information and detect identity theft --Worked in teams to understand and to communicate potential fraud rings and trends --Handled inbound and outbound calls in order to help guests with card issues and to deter fraudsters --Analyzed transaction histories and account information for possible credit card fraud --Used credit bureau software and public records to verify sensitive information and detect identity theft --Worked in teams to understand and to communicate potential fraud rings and trends
-
-
-
DK Education
-
China
-
Education Administration Programs
-
1 - 100 Employee
-
Deputy Academic Director
-
Aug 2014 - Aug 2015
--Trained new teachers and guided them in curriculum development--Gave marketing demos to parents, administration and students for recruitment purposes--Created a new, researched-based teacher-training curriculum--Directed curriculum development for American Literature and SAT test courses--Collaborated with the product and engineering teams on an online SAT training program by adding components and making it more user friendly--Completed research and published findings about taking the SAT test for company’s annual symposium Show less
-
-
Teacher
-
Aug 2013 - Aug 2014
--Taught American Literature and the SAT test to middle school, high school and college students--Collaborated with team members to design a middle school writing, literature, and speaking course--Received and gave feedback to fellow teachers--Proofread and graded student essays
-
-
-
College of Saint Benedict and Saint John’s University
-
United States
-
Higher Education
-
400 - 500 Employee
-
Teaching Intern
-
Aug 2012 - Dec 2012
● Taught psychology lab classes to undergraduate students ● Developed and executed lesson plans ● Responsible for grading and maintaining database of students ● Incorporated technology and video examples to help illustrate psychological concepts ● Reviewed and analyzed personal teaching videos with supervising professor ● Taught psychology lab classes to undergraduate students ● Developed and executed lesson plans ● Responsible for grading and maintaining database of students ● Incorporated technology and video examples to help illustrate psychological concepts ● Reviewed and analyzed personal teaching videos with supervising professor
-
-
Education
-
2009 - 2013
St. John's University
Psychology, English
Community
