PV

Prithu Verma

Data Engineering & Research

Computer Science graduate student with experience in Python, SQL, machine learning, and data engineering, building end-to-end ETL pipelines, relational databases, analytical dashboards, and ML systems. Experienced in working with large-scale datasets, statistical analysis, feature engineering, and unsupervised learning, with a focus on turning complex data into measurable insights.

📍 New Delhi, India

Latest
  • Submitted my paper on anomaly detection in database query logs to IEEE INDICON 2026
  • Co-authored paper on LLM causal reasoning under review at ACL ARR 2026
  • New project: analysing 12 million companies with PySpark on Databricks
Prithu Verma

Education

  1. SRM Institute of Technology

    Expected July 2027

    B.Tech Computer Science and Engineering

    Delhi NCR, India

    Affiliations

    • IEEE Computer Society
    • SRM-ACM
    • TedxSRM

    Coursework

    • AlgorithmsA+
    • Artificial IntelligenceA+
    • Natural Language ProcessingA+
    • Linear AlgebraA+
    • Data StructuresA
    • Operating SystemsA
    • Database SystemsA
    • Computer NetworksA
    • Discrete MathematicsA
    • Probability & StatisticsA

Research

Papers and preprints. Currently submitted, under review or published

  1. A Hybrid Unsupervised Learning Framework for Multi-Type Anomaly Detection in Database Query Logs

    Verma, P., Verma, A.

    Under Review IEEE IndiCon 2026

  2. Multi-Dimensional Evaluatation of Large Language Models for Causal Reasoning Tasks

    Verma, A., Iqbal, F., Verma, P., Das, S, Ray Choudhury, S.

    Under review at ACL-ARR 2026

Projects

A few things I've built and written up.

  1. Company Data Analysis on Databricks

    Nearly 12 million company records from People Data Labs, processed with PySpark and Spark SQL on Databricks and presented in a Databricks dashboard: 11.7M companies across 149 industries, broken down by country and size.

    • PySpark
    • Spark SQL
    • Databricks
    • Pandas
  2. Health Insurance Claim Analysis

    750,000+ synthetic healthcare records generated with Faker, pushed through an ETL pipeline into 6 normalized MySQL tables, then queried for trends in claim costs, approval rates, providers and patient utilization.

    • Python
    • SQL
    • SQLAlchemy
    • ETL Pipeline
  3. Stock Portfolio Analytics

    A 6-table PostgreSQL schema fed by a yfinance ETL pipeline, with 15+ advanced SQL queries computing returns, volatility, Sharpe ratio and position-level P&L, surfaced through an interactive Streamlit dashboard.

    • Python
    • PostgreSQL
    • ETL Pipeline
    • Streamlit
    • yfinance
  4. Business Analysis Dashboards

    Three end-to-end Power BI dashboards across banking, toy retail and superstore data: 30+ interactive visuals, reusable DAX measures, and drill-throughs that take you from headline KPIs to the detail behind them.

    • Power BI
    • DAX
    • Power Query

Writing

Posts on things I've worked through and wanted to explain properly.

  1. BBD and Tax Loopholes

    26 September 2026 · 6 min read

    How the ultra-wealthy use Buy, Borrow, Die to grow their fortunes while paying almost no tax: borrowing against appreciating assets instead of selling, then passing them on with a stepped-up basis.

    • Finance
    • Billionaires
    • Economics
    • Wealth
  2. Stock Market Crash

    19 September 2026 · 4 min read

    What actually causes a stock market crash, from recessions and inflation to bubbles and panic selling, and the indicators that tend to show up before one.

    • Recession
    • Finance
    • Stock Market Crash
    • Stock Market
    • Economics
  3. Game Theory: Backward Induction and the Endgame Problem

    19 September 2026 · 2 min read

    Reasoning backwards from the last move, and why cooperation collapses all at once the moment players know exactly when a repeated game ends.

    • Strategy
    • Game Theory
    • Human Behavior
  4. Bonds and Debentures

    18 September 2026 · 5 min read

    How bonds and debentures work: face value, coupons, maturity, premiums and discounts, and how the two differ in security, risk and liquidity.

    • Finance
    • Money
    • Investing
  5. Game Theory: Tit-for-tat

    8 September 2026 · 2 min read

    Why repeating the Prisoner's Dilemma changes everything, and how a simple strategy, cooperating first and then copying your opponent, makes cooperation last.

    • Strategic Thinking
    • Cooperation
    • Game Theory
    • Human Behavior

Skills

Tools I reach for most often, shaded by how well I know them.

  • Proficient
  • Working knowledge
  • Basic familiarity

Languages

  • Python
  • SQL
  • Bash
  • C/C++
  • Java
  • HTML/CSS
  • JavaScript

Data

  • PostgreSQL
  • ETL Pipeline
  • PySpark
  • Power BI
  • DAX
  • Power Query
  • MySQL
  • MongoDB
  • Databricks

ML / AI

  • NumPy
  • Pandas
  • Matplotlib
  • Seaborn
  • Scikit-learn
  • FAISS
  • PyTorch
  • TensorFlow
  • DBSCAN

Tools

  • Docker
  • Terraform
  • Ansible
  • Git
  • GitHub Actions
  • Weights & Biases
  • MLflow

Certifications

Courses and credentials I've completed.

  1. Amazon Simple Storage Service (Amazon S3)

    1 May 2026

    KodeKloud

    Credential ID: d668a328-1468-4f7c-a535-4cefcadd89bb

  2. Generative AI Mastermind

    Outskill

  3. Complete Data Analyst Bootcamp from Beginner to Advanced

    1 August 2025

    Udemy

    Credential ID: UC-19bfb0bb-88ed-4ecc-a44f-66d040c49863

  4. Deloitte Australia - Data Analytics Job Simulation

    1 February 2025

    Forage

    Credential ID: JcDrXDZPL3dfNo5Fb

  5. Introduction to Programming Using Python

    1 September 2024

    HackerRank

  6. Python Masterclass

    1 August 2024

    Udemy

    Credential ID: UC-70736e0c-798a-48e8-945e-540b30919cb9

  7. Desktop and Mobile UX/UI Design

    1 August 2024

    Udemy

    Credential ID: UC-20c5f704-946c-4db4-92d0-5ab05387d81a

  8. Basics of Financial Planning

    1 July 2023

    INDmoney

    Credential ID: 7661777_69

  9. Fundamental Analysis: Understanding Financial Statements

    1 June 2023

    Udemy

    Credential ID: UC-dbfaf132-8fc6-4a0a-a3f2-77d50f3d024e

Contact

Happy to talk about research, collaborations, or interesting problems.