Brian Balagot
Data Scientist
650-837-1776
Summary

I am a Data Scientist with an MBA and a background in Applied Mathematics. I am passionate about companies that are making necessities - food, healthcare, livelihood, shelter, and education - more accessible to people. I have eight years of experience interfacing with teams from different functions and gathering requirements, using data to identify insights and build business cases, and execute and evaluate company projects.

Education
Galvanize (fka Zipfian Academy)
Data Science Immersive 2016
700+ hours of full-time and industry-focused training covering statistics, machine learning, software engineering, and working with data at scale.
Individual capstone project was BotBoosted: BotBoosted is an app prototype that explores the level of fabrication in a topic on Twitter. With fabrication referring to whether a tweet was made by a spambot or fake account, the app downloads tweets for a queried topic, classifies each tweet as human or not based on the behavior of the twitter user using an ensemble of three random forests, breaks down the tweets into the different subtopics with on the fly topic modeling using a novel heuristic algorithm named Incremental Pareto NMF, and summarizes each topic with exemplary real and fake tweets chosen with corpus summarization NLP techniques.
Tools used: Python (pandas, numpy, scikit-learn, statsmodels, scipy, matplotlib, nltk, beautifulsoup), SQL, Linux, Git, MongoDB, AWS, Hadoop, MapReduce, Spark
Curriculum: http://www.galvanize.com/courses/data-science/
Ateneo Graduate School of Business
MBA Regis Program 2015
An accelerated honors program in partnership with Regis University in Denver, Colorado. Studied each discipline: Finance, Accounting, Operations, Marketing, Quantitative Management, Change Management, Project Management, and Strategic Management with the perspective of a decision-maker.
Individual capstone project was a strategic analysis and plan for CVC Supermarket. Quantitative models used to implement the plan included product assortment optimization by measuring product substitution and incremental impact on basket size, store layout optimization by looking at category co-purchases modeled using the deltahedron technique, store location selection by modeling town/city/municipal affluence using census/population data, and distribution center location identification using the center of gravity approach
Ateneo de Manila University
B.S. Management Engineering (Applied Mathematics) 2008
An honors program focused on understanding the different management disciplines from a quantitative approach. Core courses included Statistical Analysis, Operations Research, Financial Management, Strategic Management, and Operations Management
Skills
Data Science: Natural Language Processing, Recommendation Systems, Exploratory Data Analysis, Supervised Machine Learning, Unsupervised Machine Learning, A/B Testing, Multi-arm-bandit, Statistical Analysis, Business Research Methods, Python, Pandas, NumPy, Matplotlib, Scipy, Scikit-learn, NLTK, imblearn, Unix, SQL, MongoDB, Git, Github, Microsoft Excel, SPSS, Data Munging, Regression, Regularized Regression, Document Summarization, Topic Modeling
Management: Marketing, Operations Research, Operations Workflow Analysis, Strategic Management, Finance, Human Resources, Project Management, Change Management, Product Design and Prototyping
Work Experience
Human Nature (Gandang Kalikasan, Inc.), Data Science Consultant
Dec 2016 - Jan 2017

Human Nature (http://humanheartnature.com/) is a sustainable beauty and personal care product manufacturer in the Philippines focusing on bringing "green" products that have no harmful chemicals to consumers. It is a 500 employee operations selling to multiple markets in Asia and the United States. It has its own R&D and production facilities in the Philippines, and sources materials from indigenous farmers in order to bring them out of poverty.


> A fixed engagement consultancy to study the company's web sales data and identify key insights that can be acted on to improve the relationthip with the customer

> Identified key insights regarding online customer behavior drivers that are strongly related to higher customer retention and future sales potential

> Identified major website customer clusters to guide marketing communications

> Developed a production-ready user:user recommendation system for personalized product suggestions for online customers and experiments to test multiple recommendation systems

> Studied different item:item similarity recommendation systems to identify products that can appear on the page of a product in order to suggest similar products to recommend to the user

CVC Supermarket, Business Development Manager, 
Metro Manila, Philippines
Jan 2011 - Jun 2016

CVC Supermarket is a horizontally integrated food retail supply chain of production, logistics/warehousing, and retail. A 10 store operation with 300 employees, has its own bakery, food processing commissary, and private label repacking and warehousing, aside from its own meatshop, seafood, and produce section inside the store. Founded in 1956, it is one of the pioneer chains in the Philippines.


> Developed and launched the company’s first take-out catering product line by using market research skills, and by coordinating production with Commissary Operations, resulting in 6-7x more holiday revenue for food

> Started the company’s first B2B sales team using entrepreneurial skills and data-driven decision-making for product line selection, resulting in 800 new business accounts and additional annual revenue of Php 50M

> Developed a cost estimation model that analysed overhead cost patterns via inventory and cost usage patterns, as a guide for savings decisions and cost control projects, resulting in additional 3% net profit.

Information Technology Manager, 
Metro Manila, Philippines
May 2008 - Dec 2010

> Reengineered the product data encoding process by applying six sigma failure analysis skills leading to the reduction of product pricing errors, and minimizing the number of instances that vendors were overpaid

> Implemented the semi-automation of inventory-taking in order to cut down the time to produce annual inventory reports from 9 weeks to 2 weeks

> Implemented the Point-of-Sale system at CVC in order to being computerization to the cash registers being used and simplify document tracking, report creation/generation, and business intelligence