Skip to content
View 83Gh0st's full-sized avatar

Block or report 83Gh0st

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
83Gh0st/README.md
header banner

LinkedIn Email Repos


About

I'm an engineer working across Applied AI/ML, Industrial Data Science, and Software Engineering. During my internship at Yokogawa (Abu Dhabi), I built Python automation for oil & gas engineering workflows and predictive models for live refinery and petrochemical processes on projects involving ENOC and Borouge. Outside of industry, I take ideas from concept to deployment (full-stack products, RAG systems, and computer vision solutions), and my research in anomaly detection, representation learning, and edge AI has produced IEEE and Springer publications, including an IEEE Best Paper Award.

B.Tech in Electrical & Electronics Engineering at NIT Nagaland (CGPA 9.24/10), currently pursuing a BS in Data Science & Applications at IIT Madras.


🔭 Currently Building

Founder of ORYX Agency, providing software and AI development solutions for clients.


🔬 Research & Applied AI

Project Focus Recognition
Edge-Efficient Autoencoder for ECG Arrhythmia Detection Unsupervised ECG anomaly detection for real-time arrhythmia detection on edge/wearable devices, using Flat VAE, LSTM-VAE, and a shallow Conv1D autoencoder 🏆 IEEE Best Paper Award
AI-Based Bearing Fault Diagnosis Bearing fault diagnostics for electrical machines, progressing from classical ML → VAE → 1D-CNN with latent-space anomaly scoring and class-conditional trust scoring Published · VETOMAC 2025 (Springer), IIT Guwahati
Eagle Eyes: PPE Compliance Detection Real-time YOLO-based PPE monitoring (helmet, vest) for construction sites, exported to ONNX / TFLite / TF.js for CPU, edge, and browser deployment Published · PREMI 2025 (Springer LNCS), IIT Delhi
Synapse: RAG Health Chatbot RAG over a FAISS vector store grounding Llama-2-7B responses in WHO / Gale Encyclopedia sources, reaching 91% accuracy (up from a 65% baseline) with under 200ms retrieval Personal project

📄 Publications

  • Real-Time Detection of Personal Protective Equipment in Construction Sites Using YOLO, Sasikumar, A., Ganga, D., Ngullie, N. (PREMI 2025, Springer LNCS, IIT Delhi)
  • Lightweight and Efficient Variational Autoencoder for Arrhythmia Detection, co-authored with Dr. D. Ganga (IEEE Best Paper)
  • Two papers on AI-based bearing fault diagnosis (VETOMAC 2025, Springer Proceedings, IIT Guwahati)
  • 4 manuscripts under review: edge AI, anomaly detection, uncertainty quantification, fault classification (IEEE/Springer venues)

💼 Experience

Yokogawa, Advanced Solutions & Optimization Dept., Abu Dhabi, UAE Advanced Solutions Engineering Intern · Dec 2025 – Feb 2026

  • Designed Python automation for oil & gas well-test accept/reject analysis: rule-based logic, data-quality checks, automated reporting
  • Contributed to AI-driven ethane steam-cracking optimization (Borouge AI Optimizer): conversion-prediction models and severity-aware multivariate optimization
  • Built inferential soft-sensor models for refinery distillation on the ENOC project using Linear Regression, PCR, and PLS

Noviqa Management Services, Independent Software Engineering Consultant · 2026

  • Designed and deployed a production-ready business website (Next.js, React, TypeScript, Tailwind CSS, Vercel)
  • Engineered secure backend workflows: server-side validation, rate limiting, CSP, HSTS, DKIM/SPF email authentication
  • Managed end-to-end deployment: custom domain, DNS, SSL provisioning, business email

🎓 Education

National Institute of Technology (NIT) Nagaland, B.Tech Electrical & Electronics Engineering · CGPA 9.24/10 · 2022–2026 Indian Institute of Technology (IIT) Madras, BS Data Science & Applications (ongoing, online degree) · 2022–2028


🛠 Tech Stack

Languages Python C Java JavaScript TypeScript SQL MATLAB Bash

AI / ML TensorFlow PyTorch OpenCV Scikit-Learn Hugging Face LangChain

Web & Application Development Next.js React Node.js FastAPI MongoDB Tailwind CSS Stripe

Cloud & Tools AWS Docker Vercel Git

Industrial & Process Engineering: Time-Series Analysis · Statistical & Multivariate Modelling · Process Optimization · Soft Sensors · Automation · Control Systems · Signal Processing


🏆 Leadership & Awards

  • Best Paper Award, IEEE 2nd International Conference on Circuits, Power & Intelligent Systems (CCPIS 2025)
  • First Prize, Hackathon, NIT Nagaland (Nov 2025)
  • Secretary, Entrepreneurship & Incubation Cell (IIC), NIT Nagaland
  • Committee Member, IEEE Student Branch, NIT Nagaland

"Execution over ideas: ideas without execution are just dreams."

footer banner

Pinned Loading

  1. Eagle-eyes Eagle-eyes Public

    Eagle Eyes project implements real-time object detection using YOLO (You Only Look Once) with an ONNX, TFLite, TF.js model and OpenCV’s Deep Neural Network (DNN) module. It enables efficient detect…

    Jupyter Notebook 3

  2. quizo quizo Public

    Quizo is a web-based quiz application designed to be customizable and user-friendly. With Quizo, you can quickly build and deploy your own quizzes. Whether you're creating a personal project or a c…

    TypeScript 1

  3. Synapse Synapse Public

    Synapse is an advanced AI-driven medical chatbot that leverages a Retrieval-Augmented Generation (RAG) approach combined with a vector database to provide accurate, context-aware responses for heal…

    Python 1

  4. Pegasus Pegasus Public

    Designed for workplace safety and compliance monitoring, Pegasus is an ideal solution for applications in construction sites, factories, warehouses, and industrial facilities. The model ensures rea…

    Jupyter Notebook 1

  5. Vimra Vimra Public

    Vimra is a comprehensive course generation application designed to streamline the creation and management of educational content. Leveraging technologies such as MongoDB, Stripe, OpenAI, Prisma, Ne…

    TypeScript 1