Yaroslav Sergienko interactive animation • data visualization • full-stack • data analysis resume • print version of this portfolio (6M, 15 pages) Front-end — TypeScript — HTML5 Canvas — React.js — HTML5 & CSS3 — d3.js — jQuery — Ramda.js — Mobile-friendly — WebGL, Three.js Python — advanced matplotlib — pandas — cairo — numpy — Jupyter internals — Tornado — Sphinx (documentation) — scikit-learn — Selenium web driver — web scraping — Google App Engine — SQLite Other — Emacs — Arch Linux — LaTeX — Bash — SSH — Gephi — Processing.org — QGIS Office communications analysis July – August 2017 Source of web apps — The project has been completed during internship atHabidatum . — I've made a lot of data analysis, static and interactive visualizations. — The data includes private and public messages, commits and issues.
15
Embed
Yaroslav Sergienko - pallada-92.github.io · Yaroslav Sergienko interactive animation • data visualization • full-stack • data analysis resume • print version of this portfolio
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Yaroslav Sergienkointeractive animation • data visualization • full-stack • data analysis
resume • print version of this portfolio (6M, 15 pages)
Front-end
— TypeScript
— HTML5 Canvas
— React.js
— HTML5 & CSS3
— d3.js
— jQuery
— Ramda.js
— Mobile-friendly
— WebGL, Three.js
Python
— advanced matplotlib
— pandas
— cairo
— numpy
— Jupyter internals
— Tornado
— Sphinx (documentation)
— scikit-learn
— Selenium web driver
— web scraping
— Google App Engine
— SQLite
Other
— Emacs
— Arch Linux
— LaTeX
— Bash
— SSH
— Gephi
— Processing.org
— QGIS
Office communications analysis July – August 2017 Source of web apps
— The project has been completedduring internship at Habidatum.
— I've made a lot of data analysis,static and interactive visualizations.
— The data includes private andpublic messages, commits andissues.
— Fully procedurally generated, noraster or vector files. This approachsignificantly reduces network usageand makes the map fullycustomizable.
— Starting with Chris Whong'sWhashington-like styled map all thekeypoints such as stations, bendsand park corners were manuallyadded.
— A seperate subsystem has beenmade for managing marks, addingconstraints and enhancing stationscenters using simple imageprocessing algorithm.
— System support constraints suchas "angle between theese pointsmust be 45°", "theese two linesmust be parallel, and a distancemust be equal to the line widthmultiplied by 1.2".
— This enabled lines to be thinner,animations to be added, providedcustom station design, andenhanced the speed and size of themap.
— A set of utilities to extract online petfood shop offers, and to matchoffers with other offers for thesame product.
— The problem was there were a lotof variations in the how the sameproduct was named, and often thetitle was incomplete.
— The solution was to manuallycreate a decision tree and to try todeduce missing characteristics ofan offer, if possible.
Geometry problems semantictree builder July 2016
Russian passport numbercorrectness checker December 2016 live version
— It was challenging to reverse-engineer the way Russian passports are assignedto numbers using indirect data such as numbers of stolen or expired passports.