Pramika Garg Software Engineer · Researcher · Builder

Pramika Garg

Software Engineer · Researcher · Builder

About

I'm heading to Columbia this fall for my Master's in Computer Science, specializing in systems and distributed computing. This builds on two years as a Software Engineer at Qualcomm, where I've architected microservice validation platforms, cut regression cycles from 7 days to 2, and built performance benchmarking infrastructure that improved platform throughput 5×. I'm looking for a Summer 2027 internship where I can keep building systems that hold up under real load.

Outside of work, I've published six peer-reviewed papers, done federated learning research at EPFL, and built things like a fault-tolerant key-value store inspired by Amazon Dynamo and a multi-agent LLM system for automated bug triage.

0 Research Papers

0 Citations

0 Scholarships

0 Awards

Education

Columbia University

Master of Science in Computer Science

New York, United States of America Sep 2026 — Dec 2027 (Expected)

Vellore Institute of Technology

B.Tech in Computer Science & Engineering

Vellore, India 2020 — 2024 GPA: 8.8/10

Merit Scholar Top 10 / 2,100 · 2 years

Executive Board ACM / ACM-W VIT

Senior Technical Committee IEEE VIT

Technical Committee MFC VIT

Experience

Qualcomm Full-time

Nov 2025 — Jul 2026

Software Engineer I Hyderabad, India

Architected a scalable microservice validation platform using Java, Docker, and Kubernetes, re-engineering Jenkins CI/CD pipelines to reduce regression cycles from 7 days to 2 days while expanding automated coverage by 80%.
Engineered distributed performance benchmarking infrastructure with Locust, Python, and Kafka, simulating high-concurrency IoT telemetry workloads that reduced infrastructure costs by 50% while improving platform throughput 5×.
Owned end-to-end validation for IoT, AI/ML, and cloud platform releases, partnering with SWEs to investigate and resolve 100+ production-critical defects before customer deployment.
Collaborated across platform, infrastructure, and machine learning teams to drive release readiness, establishing validation strategy during high-priority release cycles.

Qualcomm Full-time

Jul 2024 — Oct 2025

Associate Software Engineer Hyderabad, India

Engineered automated data verification microservices using SpringBoot across AWS RDS and Cassandra, eliminating manual database triage and accelerating fault analysis by 50%.
Built backend automation services in Python for mission-critical workflows including license activation, IP validation, and telemetry, eliminating repetitive manual validation and improving release quality across multi-tenant public APIs.
Designed and shipped "Auto Test Analyzer," an AI-driven developer tool that diffs regression execution reports and synthesizes edge-case datasets for AI/ML inference pipelines; recognized with Qualcomm's Orion Award for engineering impact.

GE Aerospace Internship

Jan 2024 — Jun 2024

Software Development Engineer Intern Bengaluru, India

Engineered a dynamic Playwright based automation framework for the Engine Change Tool, covering 40+ complex flows; implemented logic to validate variable resultant data states, ensuring stability across diverse fleet scenarios.
Optimized asset transfer workflows by eliminating manual validation loops, resulting in a ~95% reduction in processing errors and saving 583 man-hours annually for the support team.
Collaborated with cross-functional QA teams to document workflows and validation logic, facilitating smooth knowledge transfer and accelerating feature delivery within an Agile environment.

Expedia Group Internship

May 2023 — Jul 2023

Software Development Intern - Flights Gurgaon, India

Built from scratch a configurable Maestro-based framework with 30+ toggles, covering the entire scope of user touch story for Flights LOB.
Developed the complete suite of UI and Flow-based Test Cases for Flights, later turned into a Synthetic test suite to monitor API downtime and unintentional UI changes.
Scaled solution across 4 Business Units, enabling 5X faster test case development and auto regression testing via GitHub Actions.
Worked on iOS and Android codebases to build Accessibility Identifiers, making UI elements more accessible for test automation.

EPFL (Swiss Federal Institute of Technology) Research

May 2022 — Jul 2022

Summer Research Scholar — SaCS Lab Lausanne, Switzerland

Selected among top 2% of global applicants for fully funded fellowship to conduct federated machine learning experiments focused on distributed deep learning under resource constraints.
Designed training loops and simulation pipelines to evaluate model performance on heterogeneous data; implemented lightweight pre-processing to mitigate bandwidth bottlenecks.
Optimized aggregation strategies and communication frequency, achieving 18% improvement in convergence speed and reducing mean network communication overhead by 23.2%.

The D. E. Shaw Group Fellowship

Oct 2021 — May 2022

DESIS Ascend Educare Fellow Remote

Selected as one of 40 fellows from 6,200+ applicants (Top 0.65%) for exclusive 6-month mentorship program; received weekly technical guidance from senior DESIS engineers.
Developed a fully fledged Recruitment CMS in Node.js, designed to streamline candidate tracking and automate application workflows.
Led a team to architect Fin€fy, a novel fintech product designed to revolutionize financial independence for homemakers; built the backend infrastructure and presented the MVP to firm leadership.

Research

6 double-blind peer-reviewed publications across IEEE and Springer, with 11+ citations.

Harnessing Creative Methods for EEG Feature Extraction and Modeling in Neurological Disorder Diagnoses

Randomised Analysis of Backtracking-based Search Algorithms in Elucidating Sudoku Puzzles Using a Dual Serial/Parallel Approach

Enhancing Visibility: Multiresolution Dark Channel Prior for Dehazing and Fog Removal in Images

An Exhaustive Comparative Analytical Study of 15 Machine Learning Models for Automated Cardiovascular Disease Classification

Inventory Management in AR Using Apple ARKit for VisionOS

Leveraging BBTS and NFTs for Ethical Ticketing in Live Events

Awards & Scholarships

Orion

Qualcomm

Orion Award

Led to early promotion to Software Engineer I

2025

Generation Scholarship (APAC)

$1,000 USD · Selected 180 / 20,000+ applicants

Merit Scholarship

2 consecutive years

Summer Research Scholarship

3,600 CHF for federated ML research

Harvard University

WECode Scholarship

Selected 250 / 5,000+ applicants

#6 / 1,385

American Express

AI/ML Hackathon

Individual track

Oct 2021

Top 0.65%

The D. E. Shaw Group

DESIS Ascend Educare Scholarship

Awarded INR 50,000 stipend

Oct 2021

Cloud Native Scholarship

Selected 325 / 15,000+ applicants

Projects

Python LangGraph Whisper

Adversarial Presentation Prep Agent

Local Multi-Agent Presentation Coach

Built a fully local multi-agent presentation coach using LangGraph and Ollama that conducts adversarial mock defenses across 8 attack strategies, simulating committee-style evaluations.
Implemented real-time speech evaluation using Whisper, librosa, and praat-parselmouth to score answers on content, confidence, pacing, and delivery.

Java Distributed Systems Consistent Hashing

Mini Dynamo

Fault-Tolerant Distributed Key-Value Store

Implemented a fault-tolerant distributed key-value store inspired by Amazon Dynamo, supporting partitioning, replication, and decentralized request routing.
Built a consistent-hashing ring architecture with quorum-style coordination and recovery mechanisms to restore cluster consistency after node failures.

Python LangGraph RAG

Multi-Agent Bug Triage

Automated GitHub Issue Triage System

Developed a collaborative multi-agent system that analyzes GitHub issues and routes bugs to the most relevant subsystem or owner using specialized LLM agents.
Integrated retrieval-augmented reasoning over historical issues and repository context to improve classification accuracy and reduce manual triage effort.

TypeScript VS Code API AST Analysis

VSCode Dead JSON Field Detector

Static Analysis VS Code Extension

Built a static-analysis extension for Visual Studio Code that detects unused or orphaned JSON configuration fields across large codebases.
Implemented AST-based dependency analysis and project-wide symbol resolution to identify dead configuration keys with minimal false positives.

Python 3 POSIX Shell

Parsify

POSIX Shell Token Parser

Implemented a parser for the POSIX.1 Shell Command Language specification, supporting shell token recognition as defined by IEEE Std 1003.1-2017.
Designed standards-compliant parsing logic for shell grammar constructs while preserving compatibility with the unified POSIX Shell and Utilities interface.

Technical Skills

Languages

Java Python Swift JavaScript SQL LaTeX

Backend & Data

SpringBoot FastAPI Node.js RabbitMQ Kafka MySQL PostgreSQL

Cloud & DevOps

AWS GCP Kubernetes Docker EKS Terraform CI/CD

Tools & Platforms

Git Linux Splunk Grafana XCode Jupyter Tableau

AI/ML

TensorFlow OpenCV Federated ML LLMs Data Analytics

Mobile & Testing

iOS SwiftUI UI Testing Maestro Playwright Nightwatch.js Automation

Certifications

Software Engineering Virtual Experience JP Morgan Chase

Student Expert Postman

Engineering Virtual Program Goldman Sachs

Fine Tune BERT for Text Classification with TensorFlow Coursera

Deploy to Kubernetes in Google Cloud Google

Build and Secure Networks in Google Cloud Google

Insights from Data with BigQuery Google

Perform Foundational Data, ML, and AI Tasks in Google Cloud Google

Integrate with Machine Learning APIs Google

Engineer Data in Google Cloud Google

Agentic AI for Developers: Concepts and Application for Enterprises LinkedIn Learning

Generative AI: Introduction to Large Language Models LinkedIn Learning

RAG Systems O'Reilly

Let's Connect

Open to opportunities, collaborations, and conversations about distributed systems, research, or anything interesting.

pg2899@columbia.edu Resume