Keria Bermudez-Hernandez

Ph.D. Neuroscience

I’m a full-stack data scientist with experience in media and entertainment, consumer electronics, and health and life sciences. I have built data pipelines (PySpark, BigQuery, Snowflake, dbt, and Python), analytical tools (Databricks, Tableau) and machine learning models (Databricks, Python, AWS SageMaker) with the ultimate goal of assisting teams across the company. I have partnered with marketing, strategy, product teams to deliver consumer insights and drive data driven decisions.

Prior to becoming a data scientist I was a postdoctoral fellow at NYU where I worked as a researcher and built computational tools and published numerous papers in bioimage informatics and Neuroscience.

Resume

Education

  • April 2012 - January 2015

    New York University

    Ph.D.

    Neuroscience
    Completed a Ph.D. in Neuroscience from NYU.

  • August 2007 - April 2012

    New York University

    M.S.

    Neuroscience
    Completed a masters in Neuroscience from NYU .

  • August 2003 - June 2007

    University of Puerto Rico

    B.S.

    Biology
    Graduated from University of Puerto Rico, Rio Piedras, PR, with a major in Biology and honors.

Experience

  • October 2022 - June 2023

    SONOS Inc.

    Principal Data Scientist

    • Technical lead for Data Insights team; managed data scientists and Sr. Data analysts on quarterly analyses including multiroom speaker configuration, secondhand product use, and product launch order analyses. Liaison to VP of Strategic Finance and supported quarterly earnings analytic output.
    • Led and contributed to development and productionalization of forecasting models using AWS SageMaker and Python. Models consisted of weekly and monthly forecasts of new customers that were updated weekly and published in Snowflake and Tableau and implemented by Strategic Finance for scenario testing.
    • Generated and shared multiple insights on customer behavior; discovered the relationship between Sonos product launches and repurchasing metrics that resulted in the understanding of how products drive new customers and purchasing value.
    • Developed and implemented an agile framework inspired by data driven scrum to manage analytics and machine learning projects that resulted in better goal settings and continuous delivery of value.
    • Designed and documented first code review process for Data Insights team including implementation of Git feature branch workflow resulting in more reproducible and readable code.
  • January 2021 - October 2022

    SONOS Inc.

    Sr. Data Scientist

    • First in data organization to use AWS SageMaker as data platform for machine learning; created first code base (GitHub repository) and user guide that led to implementation of AWS SageMaker by three data teams.
    • Created Tableau dashboards of subscription and engagement metrics, KPIs, to track performance of free and paid Sonos Radio, the first Sonos Software-as-a-Service (SaaS) offering. Dashboards were utilized by product managers to set yearly and quarterly OKRs and to identify drivers of subscriptions and engagement resulting in data driven improvements to product roadmap.
    • Sourced and shared insights related to free and paid Sonos Radio with product managers; led to the understanding of which product features impacted product engagement, resulting in product improvements. (MS SQL, Snowflake, Tableau).
    • Created hundreds of customer features by developing a code base in dbt and Snowflake leading to insights on how Sonos customers’ behavior and purchases were associated with repurchasing rate and lifetime value; resulted in strategy updates for bundling products.
    • Developed and implemented first productionalized models (12) in Sonos data organization that generated weekly forecasts of product registrations by product category; forecasts saved and displayed in Snowflake, Tableau.
  • April 2020 - November 2020

    Quibi Inc.

    Sr. Data Scientist

    • Generated thousands of engagement and subscriber metrics by developing a flexible data pipeline in BigQuery and dbt utilized by data science team to understand drivers of free trial conversion and retention.
    • Partnered with product and marketing teams in the planning, implementation, and analysis of A/B tests by identifying the right success metrics and conducting proper statistical analyses.
    • Extracted product engagement and business insights such as how customer features and behaviors were associated with free trial conversion, churn, and retention using BigQuery, Python, and Tableau and communicated those insights to marketing and product stakeholders.
  • October 2019 - February 2020

    Showtime Networks Inc.

    Sr. Data Scientist

    • Developed robust movie and episodes metadata pipeline in PySpark and Python that incorporated over 10 data sources; data utilized by programming managers.
    • Aided in the design and statistical analyses of A/B tests that assessed the effectiveness of different email marketing strategies.
    • Mentored data analyst in Python data analysis for over a year.
  • November 2017 - October 2019

    Showtime Networks Inc.

    Data Scientist

    • Built and collaborated on 5 PySpark data pipelines generating thousands of subscriber and behavioral features that served as inputs to machine learning models and reports. Implemented code optimizations that led to a 83% reduction in runtime.
    • Created re-subscriber and viewership propensity models that supported email interactions, paid social campaigns, and subscription offers to win back lapsed subscribers or retain current subscribers.
    • Collaborated with research and marketing teams by building analytical tools to calculate behavioral metrics and KPIs.
    • Organized and taught a 4-month long SQL course for employees across the company, with participation of over 15 employees.
    • Speaker at the Spark AI Summit 2019.“Data-driven transformation: leveraging big data at Showtime with Apache Spark”
  • May 2017 - August 2017

    Insight Data Science

    Insight Fellow

    • Designed a classification pipeline to predict the quality of customer service chat and email interactions called TicketFilter.
    • Implemented feature engineering of chat and email text data using natural language processing.
    • Decreased the time a manager spends assessing the quality of customer service interactions by 50%
  • May 2015 - November 2017

    NYU Langone Medical Center

    Postdoctoral Fellow

    • Worked with David Fenyo, Ph.D. developing image analytical tools in Python and ImageJ Java.
    • Worked in the development of an innovative algorithm to quantify protein-protein interactions in microscopy images.
    • Developed numerous custom algorithms in Python to segment, quantify, and classify cells, as well as track cell movement, and quantify biologically relevant metrics.
    • Instruction on microscopy image analysis and processing to graduate students and postdoctoral fellows.
  • April 2010 – April 2015

    New York University

    Graduate Research Assistant

    • Worked with Helen Scharfman, Ph.D.
    • Planned, executed, and analyzed experimental data. Studied postnatal neurogenesis and neuronal placement in mice using a novel experimental paradigm in transgenic mice and molecular biology techniques.
    • Designed and optimized a novel experimental technique to serially section a mouse brain structure approximately 6mm in length and 2mm wide. This technique doubled the amount of tissue that could be analyzed, thus reducing the number of mice needed for a given experiment by half.
    • Developed a series of tools written in JavaScript to automate the image processing of more than seven hundred images in Photoshop.
    • Created a series of user interface macros in ImageJ to automate cell quantication.
    • Transformed data for thousands of data points by writing scripts in Python.
    • Prepared a user form in VBA for Excel and conditionally formatted a mouse colony inventory, making data entry more effective.
    • Trained peers in molecular biology techniques and mouse colony maintenance.
  • August 2008 - April 2010

    New York University

    Graduate Research Assistant

    • Worked with Joseph Helpern, Ph.D., testing a new MRI technique called Diffusion Kurtosis Imaging. The goal of the study was to use Diffusion Kurtosis Imaging metrics as biomarkers of Alzheimer's disease.
    • Optimized protocol for drawing regions of interests in T1 weighted images. Analyzed MRI derived metrics.
  • September 2005- May 2007

    University of Puerto Rico

    Research Assistant

    • National Institute of Mental Health (NIMH), career in opportunities (COR) fellow. As part of the fellowship, I worked as a research assistant with Juan Carlos Jorge, Ph.D., on a project that addressed, in rats, the possible behavioral and brain morphological changes due to a neonatal exposure to fluoxetine.
    • Assisted in the planning, execution, and analysis of behavioral experiments in rats.
  • June 2006-August 2006

    Princeton University

    Research Assistant

    • Part of the Summer Undergraduate Research Program in Molecular and Quantitative and Computational Biology and worked with Samuel Wang, Ph.D.
    • Collaborated in the design and optimization of behavioral experiments in rats.
  • June 2005 – August 2005

    University of Minnesota

    Research Assistant

    • Part of the Life Sciences Summer Undergraduate Research Program (LSSURP) and worked as a research assistant for Virginia Seybold, Ph.D.
    • Conducted cell culture and immunohistochemical experiments.

Coding Skills

My interest in programming started when I took an introductory class to C and C++ in 2013. Since then, I have worked mostly in Python and Pyspark to develop data pipelines, machine learning, and analytical tools.

  • Python
  • SQL
  • Pyspark
  • DBT
  • Snowflake
  • BigQuery

Software Skills

I am proficient in numerous software applications. Here are some examples.

  • Microsoft Office Suite
  • Tableau
  • Databricks
  • AWS SageMaker
  • GraphPad Prism
  • Illustrator

More Skills

  • Analytical

    Designed anlytical tools to analyze and interpret complicated data.

  • Problem Solving

    Designed creative solutions to save time and increase accuracy.

  • Teamwork

    Developed partnerships with numerous departments and divisions.

  • Languages

    English

    Spanish

Get In Touch

Contact me for any collaborations or job opportunities by clicking on the LinkedIn and email links below