Hi, I’m Sumit Bathla, a B.Tech grad in Information Technology from GGSIPU, New Delhi. I am currently working as a Data Engineer at Amazon and have earlier worked as a Software Engineer at Cadence Design Systems. I’m extremely passionate about tech, especially data and AI/ML lately. Feel free to reach out, always open to chat about new ideas, technologies, and everything in between.

socials

experience and education

Amazon

Data Engineer

Sep 2024 - Present

Hyderabad, Telangana, India

Cadence Design Systems

Software Engineer

Nov 2023 - Aug 2024 • 10 months

Noida, Uttar Pradesh, India

Amazon

Data Engineer Intern

Jan 2023 - Jun 2023 • 6 months

Gurugram, Haryana, India

Guru Gobind Singh Indraprastha University

Bachelor of Technology • Information Technology

2019 - 2023 • 4 years

New Delhi, Delhi, India

Kendriya Vidyalaya

Class X & XII • CBSE

2007 - 2019 • 12 years

New Delhi, Delhi, India

publications and projects

  • Comprehensive Study on Dog Breed Identification Using Deep Learning
    Co-authored the study and development of an advanced ResNet-50 (CNN) based classifier to predict breed of dogs through images. The research study was presented at ICDAM conference of 2023 held at the London Metropolitan University and published in the book Proceedings of Data Analytics and Management.
  • Python CodeGen
    Developed a bigram-based Large Language Model from scratch, without using API of any pre-existing LLMs. The model takes Python code as input and generates similar code indefinitely. Deployed the model to the web using Streamlit.
  • Bank Note Authentication
    Created a machine learning model using Random Forest Classifier (RFC), achieving accuracy of up to 98.78% in authenticating banknotes with the UCI dataset. Deployed the model using streamlit to predict note authenticity in real-time based on variance, skewness, kurtosis, and entropy.
  • Exploratory Data Analysis Web App for Text Conversations
    Developed a Python web app for text conversation analysis, providing dynamic insights across eight dashboards. Utilized Matplotlib to visualize trends, word usage, and engagement patterns.
  • Spaceprob
    Created and deployed a Python package to the official Python Package Index (PyPI). The package focuses on utilizing web data to derive metrics such as distance measurements for space probes like Voyager 1 and Voyager 2.

certifications

trivia

  • I have an average typing speed of over 90 WPM! Checkout my profile on monkeytype.
  • I was among the top 10 players in Rocket League in India from 2018 to 2021. It is one of the only two esports games that were intially featured as part of the Olympics under the title Intel World Open.