SONIYA GOSAVI

2502-9981 Whalley Blvd,

Surrey, British Columbia, V3T 0G6
sgosavi00@mylangara.ca
(236) 867 1304
linkedin.com/in/soniya10/

SUMMARY: Highly motivated, result-driven IT Professional with 3.2 years of experience on SSIS, SSRS,
dynamic Stored Procedures and various machine learning models. A data analyst graduate who is
passionate about using data to drive decisions and effectively communicating actionable insights.
Having strong verbal and written communication skills and solid analytical skills. Strongly motivated
professional with proven ability to meet deadlines and a team player. The one who quickly gains the
deep understanding of company’s mission.

TECHNICAL SKILLS

Languages/APIs: C, C++, Python, R, .NET

IDEs/Software: RStudio, Eclipse, NetBeans, SQL Server, PostgreSQL, SAS, MS Office
Databases: SQL Server (PL/SQL), MongoDB, Oracle, MySQL
Exploratory Data: Univariate/Multivariate Outlier detection, Missing value imputation,
Analysis Histograms/Density estimation, EDA in Tableau
ML Proficiency: Data Cleaning, Data Wrangling, Data Exploration, Data Analysis, Data
Validation, Hypothesis Testing, Model Building
Supervised Learning: Linear/Logistic Regression, Lasso, Decision Trees, Ensemble Methods,
Random Forests, Support Vector Machines, Gradient Boosting, XGB, Deep
Neural Networks, Bayesian Learning
Unsupervised Learning: Principal Component Analysis, Factor Analysis, K-Means, Hierarchical
Clustering
Feature Selection: Stepwise, Recursive Feature Elimination, Relative Importance, Filter Methods
Statistical Tests: Chi-Square tests, Auto Correlation tests, Normality tests, Residual diagnostics,
Anova
Sampling Methods: Bootstrap sampling methods and Stratified sampling
Model Tuning/Selection: Cross Validation, AIC/BIC Criterions, Grid Search and Regularization
Visualization Tools: Tableau Desktop, Tableau Prep, Microsoft Excel, Power BI, Matplotlib,
Seaborn, ggplot2
Machine Learning Libraries: Numpy, Scipy, Pandas, TensorFlow, Theano, PyTorch, Scikit- learn, Keras,
spaCy, gensim

EDUCATION
Langara College, Vancouver, BC January 2020 - April 2021
Post Degree Diploma, Data Analytics
Dean Honors Role | Jade Volunteering Award

Capstone Project (Coursework)

• Extracted utility related tweets and performed.
o Preprocessing: Removed punctuation, stop words, duplicate tweets, hashtags, accented
characters, mentions, URLs and performing stemming and lemmatization

1|P a g e
Soniya Gosavi 234-867-1304
sgosavi00@mylangara.ca
o Filtering: Filtered tweets that contains at least one keyword from the keywords bank
provided by client
o Topic Modelling Techniques: TFIDF, LDA, BERT, BERT+LDA
o Clustering techniques: Agglomerative hierarchical clustering
o Metrics: Tweets in Topic, Total Words in Topic, Total Keywords in Topic, Average
keywords per tweet, Average keywords in topic (%) and Coherence score
o Inference: All these processes was useful for understanding the topics related to utility
and sentiments of the users for client

Pune University, India July 2012 - June 2016

Bachelor of Engineering, Computer Science

EMPLOYMENT EXPERIENCE
Specialist (Machine Learning Engineer) October 2016 – November 2019
Sagitec Solutions Pvt. Ltd, Pune
Employee Recognition Award 2018

• Trained the fraud detection models using supervised learning techniques which identify
differences and similarities between behaviors of genuine and fraud customers and predict
fraud transactions.
• Providing input as customers of same category the model predicts whether the individual has
insured may too much or way too less insurance and depending on that it suggests appropriate
insurance for an individual.
• Used supervised machine learning such as decision tree and random forest to predict the claims
activity which contributes to the improvement of pricing models and reduces the chances of
financial loss for the company.
• Developed Extract Transform Load (ETL) packages for extracting the data and loading it into the
appropriate tables in the database using SQL Server Integration Services (SSIS) and deploy the
data; scheduled the jobs to do these tasks periodically.
• Created parameter-based reports, graphical reports, drill-down reports, and tabular reports
based on the business requirements using SSRS and integrated in SharePoint environment for
the easy access of team. Also, involved in debugging, and testing of reports in SSRS.
• Gathered interpreted large datasets which contains millions of rows, analyzed the hidden
pattern, and generated plots and charts to communicate data and findings visually using
Tableau, which will help others understand the pattern and helps in accurate and successful
data management.
• Presented data and conclusions to team to upgrade them with strategies and operations so that
they can come up with better solutions.

PROJECTS

• Udacity: Delivered project on “Movie Trailer Website” which consisted of server-side code
(object-oriented Python) to store a list of movies and generated a static web page using the
same code that allows visitors to browse movies and watch the trailers.

2|P a g e
Soniya Gosavi 234-867-1304
sgosavi00@mylangara.ca
• Udacity: Delivered the “US Bikeshare Data” project which answers interesting questions about
it by computing descriptive statistics and prepared a script takes in raw input to create an
interactive experience in the terminal to present the following statistics.
• Langara College (Term 3): Developed a web bot using selenium to extract images from twitter
by dividing our tasks into multiple call flows to automate the process and used MongoDB to
store the retrieved images. Link: https://github.com/soniyagosavi10/Twitter_WebBot_Selinium
• Langara College (Term 3): Performed Factor Analysis and Corresponding Analysis to figure out
how different orientations to happiness like meaning, engagement and pleasure is related to
the adoption of sustainable consumption and people’s happiness based on their opinions on
existing resource sustainability in Vietnam.
Link: https://github.com/soniyagosavi10/Happiness_Vietnam
• Developed an android application using SDK and Eclipse that presents resources required for
Interior Designing of a house. Designed an android application using Java Programming
Language and Arduino hardware which controls all home appliances.
• Developed an end-to-end machine learning project about predicting the current selling value of
Used Cars. The model was build using Random Forest and the input parameters were Present
Price of the Car, Kms Driven, Type of Owner (First or Second), Year of the Model, Fuel Type,
Seller Type (Dealer/Individual) and Transmission (Manual or Automatic) .The accuracy was
determined depending on Root Mean Square Value.
Link: https://soniya-carprice.herokuapp.com/

CERTIFICATIONS (2009-2021)
• Python Foundation Nanodegree Program (Udacity)
• IBM Data Science Professional Certificate (Coursera)
• ORACLE certification from NIIT Institute.
• C and C++ Certification from NIIT Institute.

3|P a g e