Sappymukherjee214/virtual-mouse-keyboard
🚀 Control your PC with Hand Gestures! A futuristic AI-powered Virtual Mouse & Keyboard built using Python, OpenCV, and MediaPipe. Touchless interaction made seamless.
An aspiring Software Developer passionate about building intelligent solutions across AI, Machine Learning, Full-Stack Web Development, and DevOps.
Repositories
Sappymukherjee214/OBLIQ-Website
OBLIQ.in is an audit and compliance workflow platform for small and mid-sized CA firms. It helps firms centralize client documents, assign and track audit tasks, manage reviews and corrections, monitor pending work, and maintain a clear audit trail of every important action.
Sappymukherjee214/Sappymukherjee214
Sappymukherjee214/github-profile-trophy
🏆 Add dynamically generated GitHub Stat Trophies on your readme
Sappymukherjee214/github-readme-activity-graph
A dynamically generated activity graph to show your GitHub activities of last 31 days.
Sappymukherjee214/github-readme-stats
:zap: Dynamically generated stats for your github readmes
Sappymukherjee214/skills-introduction-to-github-1
Exercise: Introduction to GitHub
Sappymukherjee214/skills-introduction-to-github
My clone repository
Sappymukherjee214/fake-image-detection-system
A research-grade forensic framework for detecting GAN and Diffusion-based synthetic media via Hybrid Spatio-Frequency Swin Transformers and Cross-Modal Attention Fusion.
Sappymukherjee214/Noise-Resilient-Multimodal-RAG-for-Low-Quality-Inputs
Elite-tier research framework for robust Multimodal RAG under stochastic and adversarial noise. Implements Epistemic Gating (ENG), Symmetric Information Bottleneck (SIB) filtering, and Meta-Cognitive Loops to mitigate cascading hallucinations in vision-language retrieval.
Sappymukherjee214/AI-Humanizer
A research-grade AI-to-Human text transformer featuring an iterative detection-refinement loop to bypass AI detectors and ensure high-quality, undetectable human-like content.
Sappymukherjee214/Advanced-Level-Medical-Report-Insight-Generator
Sappymukherjee214/Validation-Module_AI-Viva
An advanced NLP validation engine for AI-based oral viva systems. It cleans speech transcripts, preserves Indian scripts (Hindi, Bengali, Tamil), scores keyword completeness, and fallbacks to RAG retrieval. Features Llama-Nemotron embeddings, mDeBERTa zero-shot NLI classification, local QA LLMs, and Indic-to-English translation.
Sappymukherjee214/Advanced-Financial-Fraud-Detection-System
Sappymukherjee214/TensorTonic-Solutions
My solutions to TensorTonic problems
Sappymukherjee214/Log-Analyzer
Sappymukherjee214/Saptarshi-OS
Saptarshi_OS: A high-performance "Interactive Desktop OS" portfolio demonstrating engineering maturity. Features real-time GitHub/LeetCode data orchestration, a print-optimized ATS Resume Engine, and premium GSAP-driven systems visualization designed for world-class recruiter conversion.
Sappymukherjee214/Cybersecurity-intrusion-detection
Autonomous AI-powered Cybersecurity Intrusion Detection System using Reinforcement Learning (DQN). Features a real-time explainable AI dashboard, live telemetry, and an interactive attack simulator.
Sappymukherjee214/Research-portfolio-website
Sappymukherjee214/Snake-Game-AI
A high-performance Reinforcement Learning environment where a Deep Q-Network (DQN) agent learns to master Snake in the browser. Built with TensorFlow.js and React, featuring a premium dashboard and real-time AI vision visualization.
Sappymukherjee214/Noise-Induced-Hallucination-in-VLMs
VAlign-Robust: A research framework for quantifying and mitigating semantic hallucination drift in Vision-Language Models (VLMs) under sensory degradation and adversarial noise.
Sappymukherjee214/Memorization-vs-Generalization-in-VLMs
This research project investigates whether Vision-Language Models (VLMs) like CLIP/BLIP truly "understand" compositions or simply memorize patterns. It features a sophisticated diagnostic suite—including CKA, Concept Probing, and Negation Consistency—to quantify generalization gaps and visualize model attention via Grad-CAM.
Sappymukherjee214/MLOps-Pipeline-for-LLM-Deployment-IBM-Level
🚀 A production-ready MLOps pipeline for Large Language Models (LLMs) featuring real-time monitoring, statistical drift detection, circuit breaking, and a premium React observability dashboard.
Sappymukherjee214/LLM-Evaluation-and-Benchmarking-Framework
A production-grade LLM Evaluation & Benchmarking Framework for systematic model auditing. Features parallel benchmarking, fairness/bias detection, MMLU integration, and a real-time analytics dashboard powered by React and FastAPI.
Sappymukherjee214/multimodal-memory-AI-system
A research-grade neuro-episodic framework for semantic retrieval from personal multimodal data (Image, Text, Audio) featuring deep fusion re-ranking and graph-augmented associative neural recall.
Sappymukherjee214/RAG-Driven-Digital-Twin-for-Intelligent-Environments-XR-AI
A frontier research-grade Digital Twin architecture integrating Multi-Agent RAG with Epistemic Scene Graphs (ESG) for spatially-grounded industrial reasoning and causal counterfactual simulation. Built with Next.js, Three.js, and FastAPI.
Sappymukherjee214/Personalized-Memory-Aware-RAG-System-with-Context-Evolution
A tier-1 research-grade RAG framework featuring cognitive memory disentanglement (STM/LTM), adaptive context evolution, and autonomous intent trajectory modeling for high-stakes longitudinal AI research.
Sappymukherjee214/Effect-of-data-noise-on-LLM-learning
This project quantifies the geometric and structural "break-point" of LLM representations under data corruption. Using Intrinsic Dimension and CKA mapping, it identifies the exact transformer layers where noise destroys semantic alignment and triggers Uncertainty Calibration collapse.
Sappymukherjee214/Nano-Banana-AI
Nano banana image generator website
Sappymukherjee214/SOUL_SENSE_EXAM
A questionnare that focuses on user's emotional health and store their results