Felix GV

Summary

Experience

• Staff Software Engineer

  LinkedIn

  January 2017 – Present (1 year 3 months)

  Killing Voldemort.

• Senior Software Engineer

  LinkedIn

  January 2015 – December 2016 (2 years)

  Major contributor to Voldemort and Venice.
  
  Venice:
  
  • Venice is a new derived data key-value store providing high throughput ingestion from Hadoop and Samza.
  • Venice improves upon Voldemort in several ways, including the fact that only a single replica of data needs to be pushed across datacenters, and pushes do not need to be interrupted during (both planned and unplanned) downtime.
  • Participated in every step of the dev lifecycle: requirements, architecture, implementation, testing, deployment and partner on-boarding.
  • Primary owner of the Venice Transport Protocol, which provides full dataset checksumming and exactly-once delivery on Kafka 0.9+, way ahead of the open-source implementation.
  
  Voldemort:
  
  • Primary maintainer of the Voldemort Read-Only pipeline.
  • Improved stability, monitoring, bandwidth control and usability.
  • Made the Voldemort Build and Push jobs highly available (in the face of unplanned downtime only) and 4x more efficient on Hadoop.
  • Achieved said goals while significantly reducing code size via clean ups and refactoring.
  • Worked with the open-source community to accept their patches and provide support.

• Software Engineer

  LinkedIn

  May 2014 – December 2014 (8 months)

  Worked in the Voldemort dev team within the Data Infra org:
  
  • Leveraged Kafka's metrics code to stabilize Voldemort's quota enforcement.
  • Stabilized Faust, a Kafka consumer writing to Voldemort.
  • Participated in the requirements gathering and architecture of Venice, a new data store to replace Voldemort and Faust.
  • Mentored a great intern who worked on a Venice prototype (:

• DevOps

  Mate1.com

  May 2011 – April 2014 (3 years)Montreal, Canada Area

  Architected, implemented, deployed, scaled and operated the following new distributed system projects:
  
  • A chat cluster based on ejabberd with custom Erlang and Mnesia modules.
  • A streaming ETL pipeline based on Kafka, Avro and Scala.
  • A data warehouse system based on Hadoop, Hive, Azkaban, Camus and more.
  
  Besides learning all of the above from scratch, also worked on:
  
  • Linux, Bash, Java, MySQL, Ant, Maven, Git, Graphite, Zookeeper and more.
  • Leading open source initiatives at the company (camus2hive, camus2kafka, schema-repo-client).
  • Speccing, coordinating and reviewing work for teams of 1 to 3 colleagues.
  • Hiring and training other employees.
  
  Hisham Mardam-Bey has been my mentor through all this. I can't thank him enough.
  
  (Internship and job)

• Software Developer

  Hibe

  January 2009 – January 2010 (1 year 1 month)Montreal, Canada Area

  Analysis, documentation, design and implementation of backend functionalities and database schemas for the Hibe social network.
  
  Worked with: Groovy, Grails, MySQL, Windows.
  
  (Internship and job)

• Teacher Assistant

  École de technologie supérieure (ÉTS)

  September 2008 – December 2008 (4 months)Montreal, Canada Area

  Explanation and evaluation of the assignments for an intro to programming class.

• Programmer Analyst

  BancTec

  January 2008 – April 2008 (4 months)Montreal, Canada Area

  Development of various prototypes for the next version of the bug-tracking system used internally at BancTec. It was a requirement to use eFirst Process, which is one of their proprietary middleware product.
  
  Worked with: Java, JSP/Servlets, eFirst Process, Oracle, HTML, CSS, Windows.
  
  (Internship)

• Unified Agent

  Uniserve Communications Corporation

  August 2005 – February 2008 (2 years 7 months)

  Handling customer care and technical support calls. Troubleshooting: dial up, DSL, VoIP, web hosting, Windows, Mac.
  
  QA for the Kinzin social network project.
  
  (Part-time job)

• Webmaster

  Equiterre

  January 2006 – May 2006 (5 months)Montreal, Canada Area

  Migration of Equiterre's website from a WIMP setup (Windows, IIS, MS SQL Server, PHP) to a LAMP setup (Linux, Apache, MySQL, PHP). Development of new functionalities.
  
  (Internship)

Education

• École de technologie supérieure

  Bachelor of Engineering (B.Eng.), Software EngineeringBachelor of Engineering (B.Eng.), Software Engineering

  2007 – 2012

  Studied processes for the whole software development cycle: requirements, design and architecture, implementation, quality assurance and control, testing, deployment, maintenance, documentation.
  
  Graduated with a special recognition for significant and sustained extra-curricular involvement, which was awarded to only two individuals out of several hundreds in my cohort.
  
  Played around with: Parallel processing with LAM/MPI. OpenGL with Qt and JOGL. Java...

  Activities and Societies: Computer Science Games, Dronolab, Improv, and a bunch of other stuff...

• Collège de Maisonneuve

  Technical Degree (DEC), Software ProgrammingTechnical Degree (DEC), Software Programming

  2003 – 2007

  Many hands-on practical courses on software development. We learned about imperative and object-oriented programming, relational databases, desktop and web development, and how to read UML.
  
  The technologies studied were: Java, VB.Net, ASP.Net, Oracle and MS SQL Server.

Volunteer Experience & Causes

• Database Competition Director

  CS Games 2014 Organizing Committee

  August 2013 – March 2014 (8 months)Science and Technology

  Writing theoretical questions on relational databases as well as Big Data technologies. Preparing a practical challenge where contestants must write queries on a given dataset using Postgres. Both sections were written with the help of David Lauzon.

• Conference Organizer

  Big Data Montreal

  September 2011 – April 2014 (2 years 8 months)Science and Technology

  Organized 23 conferences over the span of ~2.5 years, then set up a team of volunteers to carry on the work. The meetup is still active and has now hosted 55+ events.
  
  https://www.meetup.com/Big-Data-Montreal/

Certifications

• Cloudera Certified Developer for Apache Hadoop (CCDH)

  Cloudera

  August 2011 – August 2011

Honors & Awards

• 3rd place in Innovative Design

  Quebec Engineering Competition

  2012

• 3rd place in Software Engineering

  Engineering Games

  2012

• 1st place in SQL Scripting

  Computer Science Games

  2011

• 1st place in a Data Visualization contest

  LOG750 class, ETS

  2011

• 3rd place in Extreme Programming

  Computer Science Games

  2010

Languages

• English

  Native or bilingual proficiency

• French

  Native or bilingual proficiency

• Spanish

  Professional working proficiency

Projects

• Voldemort

  Starting May 2014

  Voldemort is a distributed key-value store based on the Dynamo paper. It features tunable consistency, multiple storage engines, cross-colo replication and multi-tenancy at scale.

  Team members:

  • Felix GV,
  • Siddharth Singh,
  • Arunachalam Thirupathi,
  • Xu Ha,
  • Lei Gao,
  • Greg Banks

• Tehuti Metrics

  July 2014 – July 2015

  Tehuti is a simple metrics library providing statistical measurement, reporting and quota functionalities.

  Team members:

  • Felix GV

• Nagini - Distributed System Deployment Tool

  May 2015 – June 2015

  A tool to setup remote cluster, deploy and un-deploy distributed system instances (applications) and configuration on the remote cluster, bring-up or shut-down remote instances, and monitor the remote instance status.

  Team members:

  • Felix GV,
  • Xu Ha

• Schema Repo

  October 2014 – January 2015

  The Schema Repo is a RESTful web service for storing and serving mappings between schema identifiers and schema definitions.

  Team members:

  • Felix GV,
  • Ernest Mishkin

• schema-repo-client

  March 2014 – April 2014

  A strongly typed implementation of a caching client for the AVRO-1124 schema repository, written in Scala.

  Team members:

  • Felix GV,
  • Francois Langelier

• camus2kafka

  September 2013 – January 2014

  This tool lets you take Kafka topics that were previously persisted in Hadoop through Camus and push them back into Kafka.

  Team members:

  • Felix GV,
  • Boris Fersing

• camus2hive

  September 2013 – September 2013

  A simple script to automatically map new Camus partitions to Hive partitions.
  

  Team members:

  • Felix GV,
  • Jonathan Herriott

Publications

• VisReduce: Fast and responsive incremental information visualization of large datasets

  2013 IEEE International Conference on Big Data

  October 2013

  Performance and responsiveness of visual analytics sytems for exploratory data analysis of large datasets has been a long standing problem. We propose a method for incrementally computing visualizations in a distributed fashion by combining a modified MapReduce-style algorithm with a compressed columnar data store, resulting in significant improvements in performance and responsiveness for constructing commonly encountered information visualizations, e.g. bar charts, scatterplots, heat maps, cartograms and parallel coordinate plots. We compare our method with one that queries three other readily available database and data warehouse systems - PostgreSQL, Cloudera Impala and the MapReduce-based Apache Hive - in order to build visualizations. We show that our end-to-end approach allows for greater speed and guaranteed end-user responsiveness, even in the face of large, long-running queries.

  Authors:

  • Jean-François Im,
  • Felix GV

Organizations

• Big Data Montreal

  Conference Organizer

  September 2011 – April 2014

  Created and organized the Montreal-based user group for Big Data, NoSQL and distributed systems technologies.
  
  Join the group on LinkedIn: http://www.linkedin.com/groups/Big-Data-Montreal-4102009
  
  Join the group on Meetup: http://www.meetup.com/Big-Data-Montreal/
  
  Contact the current organizers: bdm-admin@googlegroups.com

Groups

• DCIETS_2013_CS_CUP

• Hadoop Users

• Apache ZooKeeper

• Big Data Montreal

• CS Games

• NoSQL

• Cloudera Hadoop Users

• See 2 more

• Erlang/OTP Professionals

• Big Data and Analytics

• See less