Skip to content
View shaun6jrome's full-sized avatar
🤨
🤨

Block or report shaun6jrome

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
shaun6jrome/README.md

header

Typing SVG


Profile Views Portfolio LinkedIn Email


〔 IDENTITY 〕

#!/usr/bin/env python3
# >> shaun_jerome.py

class ShaunJerome:

    name           = "Shaun Jerome"
    location       = "Coimbatore, TN, India"
    degree         = "B.Tech Computer Engineering"
    specialization = "Cybersecurity"
    university     = "Karunya Institute of Technology & Sciences"
    sgpa           = "8.17 / 10.0"

    stack = [
        "Python", "JavaScript (ES6+)", "Java", "C",
        "React", "Next.js", "Tailwind CSS", "Framer Motion",
        "FastAPI", "Google Gemini API", "FFmpeg", "ChromaDB"
    ]

    currently_learning = [
        "Advanced Penetration Testing",
        "LLM Inference on Apple Silicon (ARM)",
        "Zero-Trust Architecture & Secure APIs"
    ]

    fun_fact = "I won SIH 2025 — a govt-level national hackathon 🏆"

    def motto(self) -> str:
        return "BUILD · SECURE · WIN"


〔 HACKATHON HALL OF FAME 〕

Achievements speak louder than code.

🥇 Smart India Hackathon 2025

WINNER — Government of India

Sep 2025 – Dec 2025

Built an offline-capable rural banking platform using USSD for keypad phone users with zero internet. Scam-prevention, fraud-reduction, full security architecture — presented to a government evaluation panel.

Solution approved by the Government of India

✅ Addresses financial inclusion for millions of unbanked users

SIH

🥉 Aurelion National Hackathon 2026

3RD PLACE — National Level

Built a DRM Anti-Piracy System — real-time licence verification, encryption at rest and in transit, tested against bypass and tampering attacks. Competed against teams from across India.

Fully functional DRM prototype delivered

✅ 3rd place among national-level competitors

Aurelion


〔 TECH ARSENAL 〕

// Languages

        



// Frontend & Frameworks

   Next.js   Tailwind   Flask



// Security & Tools

         Kali


〔 GITHUB STATS 〕


〔 STREAK 〕

GitHub Streak


〔 ACTIVITY 〕

Activity Graph


〔 CONTRIBUTION SNAKE 〕

github contribution snake

〔 FEATURED BUILDS 〕

Project Stack What it does
👁️ KAIROS Python · FastAPI · Next.js · Gemini API · SQLite AI behavioral monitoring and drift detection system. Automated LLM red-teaming.
📊 RAG-Eval FastAPI · Next.js · ChromaDB · ONNX · Llama-3 Observable RAG pipeline — scores its own answers for hallucination. Local embeddings.
✂️ Cutwise React · Next.js · Flask · Gemini API · FFmpeg AI creator analytics — hook strength, pacing, retention, audio quality. PDF reports. Solo built end-to-end.
🏦 Rural Banking App USSD · Python · Security Engineering SIH 2025 Winner — offline banking for keypad phones. No internet. Govt of India approved.
🔐 DRM Anti-Piracy System Encryption · Licence Verification · DRM 3rd place nationally — real-time licence auth, tamper-tested, hackathon built.
🔬 Apple Silicon LLM Research ARM · LLM Inference · Quantization Independent research on LLM thermal & performance on fanless Apple Silicon.

〔 EXPERIENCE 〕

👁️ KAIROS — AI Behavioral Monitoring & Drift Detection System · Solo Developer · 2026–Present

Python FastAPI Next.js SQLite Sentence Transformers Gemini API APScheduler

  • Architected an end-to-end AI observability platform to monitor production LLMs and detect model drift
  • Built a dual-engine system with a lightweight Python SDK and FastAPI backend calculating semantic drift
  • Engineered an automated Red Team Probe Suite to stress-test target LLMs daily against 50 targeted queries
  • Developed an LLM-as-a-Judge classifier that diagnoses root causes of CRITICAL drift events and triggers webhooks
🏦 Rural Digital Banking App — Smart India Hackathon 2025 · Frontend Developer · Sep–Dec 2025

USSD Python Offline-First Fraud Prevention Security Architecture

  • Built offline-capable banking via USSD for keypad phones — zero internet required
  • Designed scam-prevention and fraud-reduction for low-literacy users
  • Architected end-to-end: backend logic, USSD flows, security layers
  • Presented to Government of India panel — solution officially approved
🔐 DRM Anti-Piracy System — Aurelion National Hackathon 2026 · Security Developer · 3rd Place

DRM Encryption Real-Time Licence Verification Tamper Testing

  • Encryption at rest and in transit for protected digital content
  • Real-time licence-verification system for content authentication
  • Tested against bypass and tampering attacks
  • 3rd place nationally among competing teams across India
✂️ Cutwise — AI Creator Intelligence Platform · Solo Developer · 2026–Present

React Next.js Flask Google Gemini API FFmpeg yt-dlp Tailwind CSS Framer Motion

  • Full-stack AI platform: hook strength, pacing, retention, audio quality, visual structure analysis
  • Google Gemini multimodal AI for scene understanding and editing recommendations
  • FFmpeg pipeline for frame extraction and timeline segmentation
  • Analytics dashboard with synchronized video playback and subtitle readability scoring
  • PDF reports for creator iteration — fully solo, end-to-end
📊 RAG-Eval — Observable RAG Pipeline · Full-Stack AI Developer · 2026

FastAPI Next.js ChromaDB ONNX Llama-3 SQLite Tailwind CSS v4

  • Architected and deployed a full-stack RAG pipeline with sub-second latency
  • Engineered an "LLM-as-a-Judge" system scoring 100% of responses for faithfulness
  • Optimized cloud constraints by replacing PyTorch with local ONNX models, reducing memory footprint by >80%
  • Built a real-time observability dashboard tracking token usage, latency, and hallucination metrics
🔬 Apple Silicon LLM Efficiency Research · Independent Researcher · 2026–Present

LLM Inference ARM Apple Silicon Quantization Edge AI

  • Analysed thermal behaviour and execution on fanless Apple Silicon MacBooks
  • Documented performance across model sizes and quantization levels
  • Evaluating power efficiency trade-offs for edge LLM deployment

〔 EDUCATION 〕

Degree Institution Year Score
B.Tech Computer Engineering — Cybersecurity Karunya Institute of Technology & Sciences, Coimbatore Expected 2028 SGPA 8.17 / 10

〔 CURRENTLY IN THE LAB 〕

$ cat currently_learning.txt

🔐  Penetration Testing     →  Advanced exploitation, CTFs, red-team fundamentals
🤖  LLM on ARM              →  Quantization, thermal limits, edge inference
🌐  Secure Full-Stack       →  Zero-trust design, auth hardening, API security
☁️   Cloud Security         →  Docker, CI/CD pipelines, cloud posture management
🧠  Multimodal AI           →  Gemini API deep-dives, video understanding, embeddings

> BUILD · SECURE · WIN

Portfolio LinkedIn Gmail GitHub

footer

Pinned Loading

  1. drm-system-application drm-system-application Public

  2. RAG-EVAL RAG-EVAL Public

    TypeScript