Shashwata Roy | CS Student & Problem Solver

About Me

Background & Education

I'm a Computer Science student currently pursuing my Master's degree at Georgia Tech with the Computing Systems specialization.

My research interests include Algorithm Design, Optimization, Theoretical CS and Cryptography. I also enjoy competitive programming and have achieved top ranks in various programming contests.

Georgia Institute of Technology

Aug 2025 - Aug 2027 (Expected)

M.S. in Computer Science

Specialization: Computing Systems

GPA: 4.0/4.0

Indian Institute of Technology, Kharagpur

Dec 2021 - July 2025

B.S. in Mathematics and Computing

Minor degree: B.Tech in Computer Science & Engineering

Experience

Internship Experience

NUS

National University of Singapore

Research Intern May 2024 - July 2024

Investigated NP-complete classifications, showing some problems are provably harder than others in randomized settings
Identified NP-complete problems linearly reducible to HAMPATH variants, proving hardness lower bounds
Studied pseudorandom generators and hitting set generators to derandomize probabilistic algorithms

NP-Completeness Randomized Algorithms Derandomization

IIT

IIT Kharagpur

Research Intern May 2023 - July 2023

Built a spatial crowdsourcing task-assignment system in Python under geospatial, skill, and budget constraints
Implemented game-theoretic and conflict-aware greedy algorithms to maximize assignment satisfaction score
Modeled probabilistic worker skills via Monte Carlo simulation for uncertainty-aware assignment
Benchmarked assignment yield, runtime, and satisfaction metrics across varied parameter settings

Game Theory Spatial Crowdsourcing Python

Build

Technical Projects

MCP Server for Local LLMs

Dec 2025 - Feb 2026

GenAI, Python, SymPy, MCP

Built an MCP framework with sandboxed Python tools, granting local LLMs verified computation capabilities
Improved Qwen3-8B HumanEval accuracy from 2.5% to 63.7% using multi-turn execution and AST parsing
Improved MATH-500 accuracy by 4.0% and reduced compute latency by 32% through deterministic delegation

View repository

Image Captioning System

Feb 2024 - Apr 2024

AI/ML, Python, PyTorch, CLIP

Architected a PyTorch pipeline coupling a frozen CLIP ViT encoder with a custom Transformer decoder
Implemented multi-head cross-attention over 196 visual tokens for visually-grounded caption generation
Reached BLEU-4 0.25, METEOR 0.47, and CIDEr 3.6 using AdamW, label smoothing, and beam search

View repository

Recognition

Achievements

◆ Competitive Programming

Global Rank 1

CodeChef August Long 2023, Division 1 (5★+)

2006

CodeChef Rating

Handle: royxd123

2097

Codeforces Rating

Handle: roycf123

#50

Codeforces Edu Round 186

Div 2 out of ~15000 participants

◇ Academic Achievements

AIR 250

ICPC Amritapuri 2023

Qualifier Round

AIR 1311

JEE Advanced 2021

Among ~150,000 candidates

AIR 715

JEE Mains 2021

Among ~1 Million candidates

RMO

INMO Qualification (2019)

State Rank 20 (Maharashtra) in RMO 2018

Expertise

Technical Skills & Coursework

Competitive programming ratings: Codeforces 2097 (roycf123), CodeChef 2006 (royxd123), plus Global Rank 1 in CodeChef August Long 2023 (Division 1).

Languages

Python C++ C

Core Areas

Data Structures Algorithms Object Oriented Programming Machine Learning

CS & Security Coursework

Design & Analysis of Algorithms Online Algorithms Theory of Computation Programming Languages Deep Learning Natural Language Processing DBMS Computer Networks Applied Cryptography Information Security

Math Coursework & Tools

Linear Algebra Probability & Statistics Combinatorics Stochastic Processes Graph Theory Git GitHub LaTeX

Get in Touch

Let's Connect

I'm always interested in discussing opportunities, collaborations, or just connecting with my fellow CS enthusiasts!

Email royshashwata8@gmail.com

LinkedIn shashwata-roy-991548226

</>

GitHub RoyCoding8

☎

Phone +1 (470) 357-1664

Send me an email