sample

Pericles (Peri) Rocha

SENIOR TECHNICAL PRODUCT MANAGER


I'm a Senior Product Manager with 27 years in analytics, and a diverse set of experiences in the software industry across product engineering, sales, business programs and marketing. I've lived and worked in three continents, which helped me build a diverse cultural background.

I am also the author of "Learn Azure Synapse Data Explorer", published by Packt Publishing. You can buy my book here (available on Kindle and Paperback formats).

On my personal interests, I'm a recording artist with a deep passion for writing and performing music. My album "Doublethink" can be found in all streaming platforms. I also enjoy studying and practicing karate. I'm an avid reader.

About me



Work: Senior Product Manager at Microsoft Fabric, Synapse Warehouse

Location: Redmond, WA, USA

Education:

  • Master of Science at University of Illinois at Urbana-Champaign, with specialization in Data Science
  • MBA at Fundação Getúlio Vargas (Brazil)
  • Bachelor of Computer Science
  • Member of Tau Beta Pi for outstanding academic performance

    PERSONAL PROJECTS



    Author: Learn Azure Synapse Data Explorer

    Disciplines: Analytics, Machine learning, big data, end-to-end cloud analytics.

    A guide to building real-time analytics solutions to unlock log and telemetry data.

    This book covers the following exciting features:

  • Integrate Data Explorer pools with all other Azure Synapse services
  • Create Data Explorer pools with Azure Synapse Studio and Azure Portal
  • Ingest, analyze, and serve data to users using Azure Synapse pipelines
  • Integrate Power BI and visualize data with Synapse Studio
  • Configure Azure Machine Learning integration in Azure Synapse
  • Manage cost and troubleshoot Data Explorer pools in Synapse Analytics
  • Secure Synapse workspaces and grant access to Data Explorer pools
  • Book on Amazon: Learn Azure Synapse Data Explorer

    Code samples: GitHub


    My Kind of Music

    Disciplines: Machine learning, sentiment analysis, text mining, search engines and text retrieval

    A recommendation system that uses text mining to suggest songs to a user based on their desired mood and a few keywords. It asks the user to provide what their desired mood is from a 5-level ordinal scale (very sad, sad, neutral, happy, and very happy) and some key words, and the software recommends songs that match that user-defined sentiment and keywords.

    Demo: My Kind of Music

    Video: Video tutorial for My Kind of Music

    Source code and documentation: GitHub



    A study of housing market trends in Austin, Texas

    Disciplines: Statistical analysis (ANOVA, collinearity, multiple linear regression, others), machine learning, data cleaning

    By using historical data, we attempted to predict home prices using multiple linear regression and other methods to find the best-possible prediction model for home prices in Austin. The use of detailed historical data about property sales, the intent is to offer home buyers guidance to help them understand if the sale price for a house is within overall market expectation, helping them on the decision-making process.

    Analysis: A study of housing market trends in Austin, Texas

    Source code and documentation: GitHub


    Narrative visualization: house prices in Austin, Texas

    Disciplines: Computer graphics, data visualization

    Plotting data on web pages using D3.js and custom data.

    Demo: Data visualization demo

    Source code and documentation: GitHub


    Video stitching and processing

    Disciplines: Computational photography

    Manually stitching hundreds of photos together to create a panorama. Creating a video that projects frames onto a reference plane

    Documentation: GitHub

    Video results: Panoramic video from image frames

    Source code: can't be shared to maintain academic integrity. Please contact me if you'd like to learn more.


    Hybrid images

    Disciplines: Computational photography

    Hybrid images are static images that change in interpretation as a function of the viewing distance. The basic idea is that high frequency tends to dominate perception when it is available, but, at a distance, only the low frequency (smooth) part of the signal can be seen. By blending the high frequency portion of one image with the low-frequency portion of another, you get a hybrid image that leads to different interpretations at different distances. This is an implementation of the techniques described in the SIGGRAPH 2006 paper by Oliva, Torralba, and Schyns

    Documentation: GitHub

    Source code: can't be shared to maintain academic integrity. Please contact me if you'd like to learn more.


    Image quilting

    Disciplines: Computational photography

    Implementation of the image quilting algorithm for texture synthesis and transfer, described in this SIGGRAPH 2001 paper by Efros and Freeman. Texture synthesis is the creation of a larger texture image from a small sample. Texture transfer is giving an object the appearance of having the same texture as a sample while preserving its basic shape.

    Documentation: GitHub

    Source code: can't be shared to maintain academic integrity. Please contact me if you'd like to learn more.


    Gradient domain fusion

    Disciplines: Computational photography

    Seamlessly blend an object or texture from a source image into a target image.

    Documentation: GitHub

    Source code: can't be shared to maintain academic integrity. Please contact me if you'd like to learn more.


    Image-based lighting

    Disciplines: Computational photography

    Create HDR images from sequences of low dynamic range (LDR) images and compositing 3D models seamlessly into photographs using image-based lighting techniques.

    Documentation: GitHub

    Source code: can't be shared to maintain academic integrity. Please contact me if you'd like to learn more.