Pranav Varshney

Third Year CS and DS Student

Ann Arbor, Michigan
pvarsh@umich.edu • +1 636-675-0940 • https://linkedin.com/in/pvarshh
Pranav Varshney

About

Passionate about technology's impact on society (education, trading, networks, crime, adtech) Curious about Astronomy, Law and Psychology.

Education

Projects

Work Experience

  • -

    Software Engineer Intern @ Scale AI

    Scale AI - Remote

    Responsibilities:

    • Interning Fall 2025 on the Scale AI, Technical Advisor team
  • -

    Software Developer Intern @ Amazon Web Services

    Bellevue, Washington - On site

    Responsibilities:

    • Interning summer 2025 on the Amazon SageMaker AI and Bedrock AI, Infrastructure team
  • -

    COMPFOR 131 Instructional Assistant @ The University of Michigan

    Ann Arbor, Michigan - On site

    Responsibilities:

    • Overhauled the course's Python curriculum, adding 3 new projects and 2 new lectures to the course, increasing student engagement by 20%
    • Developed a Python script to automate assignment grading, reducing grading time by 99%
  • -

    Software Engineer Intern @ United Wholesale Mortgage

    Pontiac, Michigan - On site

    Responsibilities:

    • Piloted a change insights data pipeline in C# and SQL to populate UWM's data lake
    • Analyed and visualized data from the data lake to identify trends and insights
    • Reengineered UWM's archaic database organization to reduce data redundancy, pushing Google Cloud migration on an accelerated timeline
    • Launched a CRUD dashboard to monitor GCP costs and usage, flagging overuse and maintaining proper data handling etiquette
    • Initialized a decision tree model in Python with the Risk and Finance teams to classify written loans as success or failures, saving $20,000 in Q3 costs
  • -

    Web Developer @ The Michigan Daily

    Ann Arbor, Michigan - On site

    Responsibilities:

    • Lead the error handling and debugging processes for the website, fixing over 75 wesite errors, typos and other broken components to make user experience more enjoyable
    • Optimized website performancy by 44%, reducing page redircts and http requests, raising engagement length by 10%
    • Provided a framework for catching and fixing errors in the future, guiding new team members during onboarding
  • -

    Software Engineer Intern @ Ratna Global Technologies

    Newark, California - Remote

    Responsibilities:

    • Spearheaded the development of a car rental website for clients, using React and Node to build a responsive, user-friendly and scalable platform
    • Built a custom chatbot using OpenAI's GPT-3 to answer customer queries, automating naive customer support tasks and saving 5-10 hours of work per week
    • Designed a database schema in PostgreSQL to store user data, reducing data redundancy and improving data retrieval times by 20%
  • -

    Academic Tutor @ Kumon North America Inc.

    Chesterfield, Missouri - On site

    Responsibilities:

    • Taught students complex math concpets like calculus I, II and basic statistics, helping them succeed in school subjects and standardized tests
    • Revamped the grading system for the center, leveraging technology to automate progress tracking and develop personalized feedback for students struggling in subjects

Research Experience

  • -

    The Liu Laboratory (The University of Michigan)

    Ann Arbor, Michigan - On site

    Responsibilities:

    • Building twin platform using BatFish & Veriflow to analyze 500+ network snapshots monthly as DSLs for learning
    • Developing a framework that automates network snapshots, increasing dataset size by 250%, enhancing scalability
  • -

    The Big-DIG Laboratory

    Ann Arbor, Michigan - On site

    Responsibilities:

    • Employed a custom school sorting function in Python to classify over 75- schools based on an array of criteras, guiding lab to conclude in 2 months instead of 6
    • Demonstrated a 99.5% statistical significance test through summplemental simmulations and visual aids in Python and R
  • -

    The London Business School

    London, England - Remote

    Responsibilities:

    • Leveraged Python to pipe data from 500 contracts in a mimicked SQL database, optimizing data extraction and transformation by over 25%
    • Determined 90% confidence in distinct differences between the structors of internal contracts and external constracts using Python and R