Howdy, I'm
Nikita

A problem-solver combining Data Science and Software Engineering to solve challenging problems and build cool projects.
Get In Touch
Responsive image

About Me

Hi there! I am a Senior at Minerva University, pursuing a bachelor's in Computer Science (expected graduation is May 2023). I love building machine learning models, tinkering with Raspberry Pis and microcontrollers, and automating monkey work. Sometimes I solve meaningful problems and build useful tools for companies, and other times I work on implementing my silly ideas. I am interested in blockchain, tools for knowledge management & discovery, and developer tools.

When I am away from the screen, I usually go on hikes, play guitar, and read books. In my daily actions, I aim to be (1) curious, someone who constantly asks questions and researches new areas; (2) entrepreneurial, someone who takes a proactive stand in life and is not afraid of starting something new; and (3) mindful, someone who is aware of his choices and continuously reflects on consequences of his actions.

Skills

Languages
Frameworks
Expertise
Other Skills
Proficient
  • Python
  • Circuit- & Micropython
  • SQL
  • R
  • Pandas & Numpy
  • Matplotlib & Seaborn
  • Tensorflow, PyTorch, sklearn
  • Hugging Face Transformers, NLTK
  • Flask, Django
  • Data Structures
  • Algorithms
  • Machine Learning
  • Computer Vision, NLP
  • Causal Inference
  • Linux
  • Docker
Intermediate
  • C
  • HTML
  • CSS
  • Javascript
  • LaTeX
  • Sagemath
  • PyStan
  • Twitter & Slack APIs
  • Bayesian Inference
  • Statistics
  • Research Design
  • AWS EC2, S3
  • Virtual Machines
  • Snowflake
  • Apache Airflow

Experience

Viasat Inc.
Data Scientist
  • Built an internal sentiment analysis and reporting tool that enables teams to assess the product's real-time performance based on social sentiment. Built with Twitter & Slack APIs, Snowflake, SQL, Hugging Face Transformers, Python, Airflow, and Docker.
May 2022 - August 2022
Carlsbad, CA
University of Chicago, Globus Labs
Researcher & Data Scientist
  • Developed a Serverless Federated Learning python library for experimenting with Federated Learning on edge devices. Accepted for publication at IEEE eScience 2022. GitHub.
  • Trained and deployed a Convolutional Neural Network across a network of edge devices to enable multi-staged real-time image processing. The codebase enables building serverless Edge-to-Cloud infrastructures in Python. Documented on Github.. Paper. Poster.
June 2021 - Current
Chicago, IL
SandBox Network Inc.
Data Engineer
  • Developed a data-scraping algorithm in Python that outputs statistics about a YouTube channel’s engagement in promotions, thus helping the company decide on potential partnerships with creators
  • Managed a team of 3 engineers by managing repositories, schedules, and offering peer-review to create a full-stack application
January 2021 - April 2021
Seoul, South Korea
Sabudh Foundation
Data Analyst
  • Designed a social impact strategy that utilizes Machine Learning at the Edge to solve traffic issues in Hyderabad. Presented the data-gathering process, stages of implementation, and impact assessment method. The plan was used to improve traffic congestion in the Hyderabad city center
September 2020 - December 2020
London, UK
Vibemap
Student Researcher
  • Produced a literature review (2500 words) on the impact of gamification on building communities.
  • Designed a gamification strategy for the app using the Octalysis Framework
  • Prototyped 10 gamification elements for the app, and visualized the UI in Adobe Illustrator
September 2019 - May 2020
San Francisco, CA
View Resume

Projects

Real-Time Camera Tracker with Raspberry Pi
Created a camera that tracks any person in the room. The system detects people in the room using a Convolutional Neural Network, does adjustments to where the camera should point using PID, and turns the camera after the person by sending signals to a servo via Raspberry Pi.
Raspberry Pi Tensorflow Python
March 2022
DCGAN for Bored Apes NFT
Designed and trained a DCGAN (Deep Convolutional Generative Adversarial Network) for generating Bored Apes NFTs.
Tensorflow Keras Numpy
December 2021
ToolSwap: web application for exchanging tools
Designed and implemented the app's database in Django \& SQL. Wrote tests for the database to optimize the team's workflow. Worked in a team of 6 other engineers.
Django SQL Docker Testing
December 2021
Carbon Emissions Predictive Model
Modeled CO2 levels from 1958 to 2021 using the Mauna Loa Observatory data set. (1) Cleaned the data, (2) created Bayesian statistical models with PyStan for the overarching trend, seasonality, and noise, (3) made predictions for years 2021-2060, and (4) evaluated the model's fit using autocorrelation plots & posterior predictive checks.
Bayesian Inference Hamiltonian Monte Carlo PyStan Pandas Numpy SciPy Matplotlib
April 2021
Plagiarism Checker
Built a plagiarism checker using rolling hashing, Karp-Rabin algorithm, and hash tables.
Hash Tables Karp-Rabin Python
April 2021
Modeling Cost of Basic Goods
Modeled price variation of 10 basic goods across 7 countries using PyStan and Bayesian Framework. Inferred how product type, store type, and location affect the price.
Bayesian Inference PyStan Pandas Numpy SciPy Matplotlib
March 2021
Light-Sound Game
A simple light-sound game for training memory and attention.
JavaScript HTML CSS
March 2021
Causal Inference
Completed several projects in Causal Inference, using Genetic & Propensity Matching, Synthetic Controls, Regression Discontinuity, Fisher's Exact Test, and others.
R ggplot2 dplyr rgenoud
November 2020
Predicting Blight Ticket Payment
Cleaned the data and trained Logistic Regression and Random Forest classifiers to predict if people will pay for their blight tickets. Tuned the models using GridSearchCV and Cross-Validation to achieve AUC ROC of 0.81.
Scikit-learn Pandas Numpy Matplotlib
August 2020
COVID-19 Misleading Graphs Analysis - Article
Analyzed misleading graphs about COVID-19 and published an article in Towards Data Science: 40k reads.
Excel Medium
June 2020
Book Review Platform
Created a full-stack platform that allows users to browse through a database of over 10k books, get information, and leave reviews.
Python Flask PostgreSQL HTML CSS
May 2020

Get In Touch

I am always open for a conversation, whether you want to discuss an exciting opportunity, novel idea, or just want to chat!
Say Hello