about
With a natural inclination for technology throughout her life, Krishel began coding website layouts and designing graphics on Photoshop at the age of 9, fulfilling 1,500 design requests online for free. Krishel spent her high school years in Brooklyn Technical High School as a Chemical Engineering major, while also taking classes outside her major such as “Design & Drafting for Production,” “Introduction to Electricity,” and “Architectural Drawing.”
Krishel majored in Economics at Hunter College and was elected as Webmaster for the pre-dental club by her sophomore year. Toward the end of college she got the opportunity to take elective courses such as “Introduction to Computer Science,” “Computers and Money,” and “Introduction to Python,” all of which amplified her interest in technology.
After graduating, she explored careers that combined technology with analytical thinking and creativity – three skills that she has honed and enjoyed throughout her life. After evaluating the culmination of her past experiences, she studied web development, computer networking, and data science to further her career as a technology professional.
Technical Skills and Knowledge
Web Development: JavaScript, Ruby, Ruby on Rails, HTML, CSS, Sinatra, jQuery, React.js, Redux, Git, Heroku
Data Science: Python, Pandas, Scikit-Learn, SQL, Jupyter Notebook
Networking: IPv4, IPv6, TCP/IP, LAN, WAN, Cisco Routers, Cisco Switches
Software: Visual Studio Code, Atom, Terminal, Microsoft Office, Windows, macOS, Linux
work
An overview of Krishel's data science and web development work.
Each title links to its respective project. Datasets are available upon request.
Data Science

Analyzing borrowers’ risk of defaulting
Libraries used: Pandas, NLTK
This project is to prepare a report for a bank’s loan division. Here we investigate if a customer’s marital status and number of children has an impact on whether they will default on a loan. The bank already has some data on customers’ credit worthiness.

Determining the market value of real estate properties
Libraries used: Pandas
Here we will use an archive of sales ads for realty in St. Petersburg, Russia, and the surrounding areas to determine the market value of real estate properties. The task is to define the parameters. This will make it possible to build an automated system that is capable of detecting anomalies and fraudulent activity.

Analyzing profiltability of mobile prepaid plans
Libraries used: Pandas, Numpy, Scipy
Two prepaid plans offered by a state mobile operator are analyzed to determine which of the two plans is more profitable.

Analyzing success of games
Libraries used: Pandas, SciPy
Historical game data containing user reviews, genres, platforms, and sales will be analyzed to identify patterns that determine whether a game will succeed in 2017 or not.

Analyzing ride-sharing patterns
Libraries used: Pandas, SciPy
Ride-sharing data will be analyzed to find patterns, understand passenger preferences, and gain insight into the impact of external factors on rides.

Classifying data plans
Libraries used: Pandas, Sklearn
Different models are investigated in order to recommend the correct phone plan based on subscriber behavior with the highest possible accuracy.

Predicting customer behavior
Libraries used: Pandas, Sklearn
Data on clients’ past behavior and termination of contracts with a bank will be used to predict is a customer will leave the bank soon. A model with the maximum possible F1 score will be built.

Finding the best place for a new oil well
Libraries used: Pandas, Numpy, Sklearn
Data on oil samples from three regions is used to create a model that will help pick the region with the highest profit margin. The Bootstrapping technique is used to analyze potential profit and risks.

Predicting the amount of gold recovered from gold ore
Libraries used: Pandas, Numpy, Scipy
Data is provided on gold ore extraction and purification. A machine learning model will be built to help optimize the production and eliminate unprofitable parameters.

Creating a data transforming algorithm
Libraries used: Pandas, NumPy, Sklearn
The Sure Tomorrow insurance company wants to protect its clients’ data. The task is to develop a data transforming algorithm that would make it hard to recover personal information from the transformed data. The data will be protected in such a way that the quality of machine learning models do not suffer.

Building a model to determine used car values
Libraries used: Pandas, LightGBM, Sklearn, CatBoost
Rusty Bargain used car sales service is developing an app to attract new customers. In that app, users can quickly find out the market value of their car. We have access to historical data: technical specifications, trim versions, and prices. A model will be built to determine the value.

Predicting the number of taxi orders for the next hour
Libraries used: Pandas, LightGBM, Sklearn, CatBoost
Sweet Lift Taxi company has collected historical data on taxi orders at airports. To attract more drivers during peak hours, we need to predict the amount of taxi orders for the next hour. A model will be built for such a prediction. The RMSE metric on the test set should not be more than 48.

Detecting negative reviews
Libraries used: Pandas, NumPy, Math, Matplotlib, Seaborn, Re, Tqdm, NLTK, spaCy, LightGBM, Torch, Transformers
The Film Junky Union, a new edgy community for classic movie enthusiasts, is developing a system for filtering and categorizing movie reviews. The goal is to train a model to automatically detect negative reviews. We’ll be using a dataset of IMBD movie reviews with polarity labelling to build a model for classifying positive and negative reviews. The goal is to have an F1 score of at least 0.85.

Predicting the approximate age of a person from photographs
Libraries used: Pandas, Tensorflow, Matplotlib
Supermarket chain Good Seed is introducing a computer vision system for processing customer photos. Photofixation in the checkout area will help determine the age of customers in order to analyze purchases and offer products that may interest buyers in particular age groups and monitor clerks selling alcohol. Here we will build a model that will determine the approximate age of a person from a photograph. To help, we’ll have a set of photographs of people with their ages indicated.

Predicting the temperature of steel
Libraries used: Pandas, LightGBM, Sklearn, Catboost
In order to optimize production costs, the steel plant Steelproof decided to reduce their energy consumption at the steel processing stage. A model will be developed that will be able to predict the temperature of the metal.
Web Development
Luminance - Compare ingredients between skincare products
Technologies: Ruby on Rails, React, Redux, PostgreSQL, Jbuilder, JSON, Aphrodite-JSS
This tool compares ingredients between skincare products to aid in identifying beneficial and/or detrimental ingredients.
PlanEat - Record and browse future dining plans by restaurant and by menu item
Technologies: Ruby on Rails, RESTful API, OmniAuth, JSON, jQuery, ActiveRecord, ActiveModel Serializers
This Rails app was created for those who often discover new restaurants and foods to try and need a place to record these new discoveries for later.
BrewTeaful - Track the teas you own and discover new ones
Technologies: Sinatra, BCrypt, SQLite, CSS, content mangement system, user authentication
Brewteaful is a tool designed for tea lovers who would like an easier way to keep track of all the different kinds of tea that they own. Use this tool to record important tea brewing information such as: brewing time, brewing temperature, purchase date, and measurements. Also, discover new teas by viewing the latest submissions of fellow users.
Potatotainment - Displays currently trending entertainment based on real-time check-in data
Technologies: Ruby, Nokogiri, Web Scraping, Object Oriented Design
This Ruby gem lists all currently trending shows and movies along with their respective details based on real-time check-in data from Trakt.tv for those who are having a hard time deciding what to watch.
connect
questions, comments, or concerns? please don't hesitate to reach out.