Available for internships & collaborations

Rahul Paul

CSE (Cyber Security) undergrad at MS Ramaiah Institute of Technology and Project Intern at Samsung R&D. I build multimodal vision-language models, full-stack AI platforms, and security tooling — end to end.

9.53CGPA / 10
1821LeetCode Rating
Top 7.3%Globally
254.8MParam VLM built
01

About

I'm a third-year Computer Science & Engineering (Cyber Security) student with a 9.53 CGPA, currently a Project Intern at Samsung PRISM / Samsung R&D Institute India. My work spans deep learning research, full-stack engineering, and applied security.

I like problems where the whole stack matters — training and distilling a 254.8M-param vision-language model down to run on-device, writing a Rubik's cube solver from scratch that averages near God's number, or shipping a DNS threat-intelligence platform mapped to MITRE ATT&CK. I care about correctness, measurement, and things that actually run.

AI / ML Research Full-Stack Engineering Cyber Security Competitive Programming
02

Experience

Project Intern — Samsung PRISM

Dec 2025 — Present

Samsung R&D Institute India · Bengaluru

  • Developing EmoCapNet, a 254.8M-param multimodal vision-language model for 7-class emotion-aware image captioning — fusing ViT-Base + GPT-2 via cross-attention and VAD-FiLM conditioning with multi-task learning on a 67K-caption emotion-augmented Flickr30k corpus.
  • Distilled the teacher into a TinyCLIP-ViT + DistilGPT-2 student (112.5M params, 2.3× smaller); achieved METEOR 22.1, CIDEr 0.363, and emotion-controllability correlation r = 0.81.
  • Engineered on-device deployment via INT8 dynamic quantization (custom Conv1D→Linear rewrite), compressing ~1GB → 153MB (6.7×); built greedy KV-cache decoding from scratch (524ms/caption CPU) with an ONNX/TorchScript export path.
PyTorchViTGPT-2Knowledge DistillationINT8 QuantizationONNX
03

Projects

Smart Course Generator

May 2026 — Jun 2026

Full-stack AI learning platform that generates structured multi-module courses and streams them lesson-by-lesson over SSE, with adaptive quizzes, flashcards, and an AI-scored mock-interview mode. Custom multi-provider AI router (Gemini → Groq → OpenRouter) with per-provider circuit breakers, backoff retries, and key rotation; quality made measurable via an LLM-as-judge eval harness in CI.

ReactTypeScriptNode.jsExpressMongoDBSSEZodPlaywright

Rubik's Cube Studio

Apr 2026 — May 2026

Implemented Kociemba's two-phase algorithm from scratch in TypeScript (cubie model, coordinate reduction, pruning tables, IDA* search) — averaging ~20.6 moves (God's number is 20) at 100% correctness across 1,500+ cubes. Runs in a Web Worker to drive a stutter-free Three.js 3D animation, with webcam cube scanning via an HSV pipeline. 66 unit tests + Playwright E2E.

TypeScriptThree.jsWeb WorkersViteVitestPlaywright

DNSentinel — DNS Threat Intelligence

Jun 2026 — Jul 2026

Full-stack DNS threat-detection platform (Manifest V3 Chrome extension telemetry) classifying DGA, tunneling, exfiltration, and C2 beaconing via a 22-feature Random Forest + Isolation Forest ensemble. Adaptive risk engine (µ + k·σ thresholding) drives a SOAR auto-block/sinkhole layer mapped to MITRE ATT&CK, streamed to a SOC dashboard over SSE. Validated over 25 real DGA families with SHAP-explained PDF reports.

PythonFastAPIReactscikit-learnDockerChrome MV3
04

Publication

Springer Nature

Applications of Quantum Computing for Internet of Things

Book chapter (pp. 79–121) in Quantum Computing, Sensing and Communications for IoT · 2026

Co-authored a 43-page book chapter surveying 70+ papers on integrating Quantum Computing with IoT for security, optimization, and intelligent decision-making across Quantum ML, Post-Quantum Cryptography, and Quantum Sensing.

05

Skills

Languages

PythonC / C++TypeScriptJavaScriptSQL

AI / ML & Deep Learning

PyTorchTransformers (ViT, GPT-2)Multimodal VLMsKnowledge DistillationINT8 QuantizationONNXTorchScriptscikit-learn

GenAI & Web

LLM IntegrationRAGPrompt EngineeringMulti-provider AI RoutingReactNode.jsExpressFastAPIRESTSSEJWT / OAuth2

Data, DevOps & CS

MongoDBSQLAlchemyDockerGitGitHub ActionsJestVitestPlaywrightpytestDSADBMSComputer Networks
06

Achievements & Certifications

LeetCode — Contest Rating 1821

Top 7.32% globally across 17 rated contests. View profile ↗

Campus Ambassador — Kshitij, IIT Kharagpur

Represented and promoted Asia's largest techno-management fest on campus.

Class XII (AISSCE) — 96.33%

Bhavan's Gangabux Kanoria Vidyamandir, Kolkata.

Certifications

IBM Threat Intelligence & Hunting · IBM Security Operations in Practice · Infosys Springboard DSA · MongoDB · Google Cloud Career Launchpad.

07

Get in touch

Have an opportunity, collaboration, or just want to talk shop about ML and security? Send me a message.