Data Scientist

Contact

Seyed Nabavi

******.******@*****.***

613-***-****

Ottawa, ON

Software

Python

SQL

NoSQL

TensorFlow

Keras

Apache Spark

Hadoop

Pandas

Matplotlib

Seaborn

Bokeh

Flask

MATLAB

Linux

JavaScript

Git

Google Cloud Platform

Education

MSc, Electrical and Computer Engineering

University of Ottawa

MSc, Electronics Engineering

Azad University

Certifications

Applied Deep learning

Machine Learning

Applied Data-Science

Time Series Analysis

Complete SQL Bootcamp

Descriptive & Inferential Statistics

The Ultimate Hands-On Hadoop

Professional Statement

A Data Scientist focused on applying statistics, machine learning, deep learning and big data techniques to develop business solutions for financial, e-commerce, sales, marketing, and retail industries.

Truly passionate about my work and always eager to connect with people. My favourite stage of a project is collaborating with our cross functional teams and working with clients to help them achieve their business objectives.

Work Experience

Data Scientist, Message Hopper Company 2017 - Present

Analyze, build, tune and deploy Crypto-Currency Market Capitalization forecasting model and achieved more than 96% accuracy of the actual price by analyzing the capitalization market, Machine Learning, Deep learning, Recurrent Neural Networks(RNN).

Explored, built, tuned and deployed a Fraud Predictor and successfully achieved 96% recall score in a highly imbalanced dataset using Data Exploration, Applied Statistics, Machine Learning Algorithms, Random Forest, Logistic Regression, SVC Classifier, and KNN Classifier.

Analyzed, built, tuned and deployed an accurate success predicting model for Bank Telemarketing Campaign using Exploratory Data Analysis, Decision Trees, Logistic Regression, and SVM .

Exploited Neural Networks, Deep Learning, Deep Features, and Image Retrieval to develop and deploy a successful Visual Product Recommender System.

Research Scientist, CAD Lab, University of Ottawa 2012 - 2017

Researched, developed and deployed in production a computationally efficient multivariate statistical model (Decoupled Polynomial Chaos)to quantify uncertainties in Radio Frequency Systems and achieved 99.9% accuracy and up to 500% reduction in computation cost compared to traditional algorithms(Monte Carlo, Stochastic Testing) using Python, Pandas, SQL, MATLAB, in Ubuntu Linux environment.

Performed descriptive and inferential multivariate statistical analysis to get insights in Radio Frequency Systems and compare new machine learning algorithms with the traditional approaches.

Researched, developed and tested a hybrid mathematical model order reduction(MOR) algorithm achieving a reduction rate of 80% and 99.8% accuracy in order to reduce Memory Consumption and Computational Cost of harmonic balance simulation technique.

Software Developer, GhodsNiroo Consulting 2007 - 2012

Architected, designed, and developed responsive client facing software applications for new generation of network-based applications and web technologies for different OS in HTML5, CSS3, Python, and JavaScript, and Rest-API.

Contact this candidate