Skip to content
View sherurox's full-sized avatar
🧑‍🍳
let him cook....
🧑‍🍳
let him cook....

Highlights

  • Pro

Block or report sherurox

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
sherurox/README.md

Hi, I'm Shreyas Pradeepkumar Khandale

Profile Views


About Me

  • 🎓 MS Computer Science (AI Track) at Binghamton University, Thomas J. Watson College — Expected Dec 2026
  • 🏫 BE Computer Science and Engineering from AISSMS College of Engineering, Pune — June 2024
  • 💼 AI Systems Integrator @ Zoo Media · Quantum ML Research Assistant @ Watson College · ITS Computer Services Administrator @ Binghamton University
  • 🔬 Research: Transformer-based probabilistic surrogate models on quantum network data — 30K–60K synthetic events per experiment, 70–80% pipeline iteration time reduction
  • 🤖 Specialising in ML Engineering, LLM/RAG Systems, IoT/Edge AI, and Quantum ML
  • 🛠 Core stack: Python · TypeScript · PyTorch · FastAPI · Docker · Angular · TensorFlow.js · Pinecone · OpenSearch
  • 📄 4 published papers in Applied ML, NLP, IoT Systems, and Financial Forecasting (IJRASET)
  • 🏆 Certified: NVIDIA CUDA C/C++ · IBM Data Science · Google Cloud Data/ML Track · Google Advanced Data Analytics

Connect with Me

Gmail LinkedIn GitHub ORCID ResearchGate Google Scholar


💼 What I'm Currently Working On

Role Organisation Focus
AI Systems Integrator (Intern) Zoo Media Inc — Remote LLM integration, prompt engineering, adaptive educational AI
Research Assistant Binghamton University (Watson College) Transformer surrogate models on quantum network data
Computer Services Administrator ITS, Binghamton University AI conversational system for ~24,000 students and staff

🚀 Flagship Projects

🧠 AI / Machine Learning

Project What it does Stack
Motion-Flow Production in-browser bidirectional sign language translation — 40+ languages, 543 landmarks/frame, sub-second latency on WebGPU Angular 21 · TensorFlow.js · MediaPipe · Three.js · Web Workers
Cashflow Management 13-week cashflow forecasting + grounded Q&A with accounting invariant enforcement Python · FastAPI · XGBoost · Pinecone · Gemini · MongoDB · Docker
arXiv Paper Curator Production RAG system — hybrid BM25 + vector retrieval, fully containerised FastAPI · OpenSearch · PostgreSQL · Redis · Airflow · Ollama · Docker
CallQA Automated voice QA harness — 10 patient personas, Groq Whisper STT, structured bug reports Python · Twilio · TwiML · Groq Whisper
Bayes Error Estimation Reproduces and extends Ishida et al. (ICLR 2023) — annotation bias, calibration, temperature scaling Python · PyTorch · NumPy · SciPy
Advanced Wine Quality Stacked ensemble (XGBoost + LightGBM + CatBoost) + SHAP interpretability + FastAPI deployment Scikit-Learn · XGBoost · LightGBM · CatBoost · SHAP · FastAPI

🌐 IoT / Embedded Systems

Project What it does Stack
Smart Bicycle Theft Prevention NFC authentication + LoRaWAN tracking + Flask web dashboard Heltec LoRa 32 · nRF52840 · LoRaWAN · NFC · Flask · TTN
IoT Network Attached Storage Sub-$100 private NAS — 300–400 MB/s SSD, cross-platform SMB/CIFS access Raspberry Pi 4 · OpenMediaVault · Linux · Bash · ext4

🛠 Tech Stack

Languages
ML & Backend
Frontend
Infrastructure


📄 Research Publications

# Title Venue DOI
1 IoT-Based Network Attached Storage — Raspberry Pi 4 NAS architecture with benchmarked USB 3.0 throughput and SMB/CIFS multi-device access IJRASET Vol.12, Nov 2024 10.22214/ijraset.2024.65616
2 Predicting Credit Card Defaults with Machine Learning — Logistic Regression vs Naive Bayes on 30,000-record Kaggle dataset with class imbalance analysis IJRASET Vol.11, Oct 2023 10.22214/ijraset.2023.55934
3 Stock Market Analysis Using Heuristic Approach — 2-layer LSTM (128→64 units) on S&P 500 (2000–2020), temporal split methodology, ARIMA/Prophet baselines IJRASET Vol.11, Oct 2023 10.22214/ijraset.2023.55932
4 Amazon Fine Food Review Analysis — TF-IDF + PCA + LDA topic modelling on 568,454 reviews, 5 interpretable themes identified IJRASET Vol.11, Oct 2023 10.22214/ijraset.2023.55930

🏆 Certifications & Credentials


📊 GitHub Stats


Building at the intersection of production ML, quantum computing, and accessible AI. 🌍

Pinned Loading

  1. Smart-Bicycle-Theft-Prevention-System Smart-Bicycle-Theft-Prevention-System Public

    An IoT-based bicycle security solution integrating NFC authentication, LoRa communication, and a web interface to provide real-time alerts and remote monitoring against theft.

    C++

  2. IOT-Based-Network-Attached-Storage IOT-Based-Network-Attached-Storage Public

    With the exponential rise of Internet of Things (IoT) devices, efficient data storage, management, and security have become critical. This project presents an IoT-Based Network Attached Storage (Io…

    1

  3. Notes-for-Hands-On-Machine-Learning-with-Scikit-Learn-and-TensorFlow-book-by-Aurlien-Gron Notes-for-Hands-On-Machine-Learning-with-Scikit-Learn-and-TensorFlow-book-by-Aurlien-Gron Public

    This repo is my chapter-wise study notebook for Hands-On Machine Learning with Scikit-Learn and TensorFlow (Aurélien Géron).

    Jupyter Notebook 2

  4. arxiv-paper-curator-rag arxiv-paper-curator-rag Public

    A production-grade Retrieval-Augmented Generation (RAG) system that automatically ingests arXiv CS.AI research papers, indexes them with hybrid search, and answers research questions using a local …

    Python