A complete machine-learning system that predicts AI assistant user satisfaction using behavioral signals such as device, usage category, time features, session metrics, and model metadata. Includes full ML pipeline, SHAP explainability, evaluation suite, and an interactive Streamlit analytics dashboard.

This is a systemC-AMS/systemC SerDes modeling Project

Interaction-centric behavioral modeling for earnings calls using multimodal neural fusion, text-audio divergence, and Q&A interaction dynamics.

Autonomous AI agent with dynamic mood-state engine, persistent user memory via entity extraction, and behavioral pacing - deployed with Docker + PM2

This repository contains Verilog HDL implementations of Half Adders, Full Adders, and 4-bit Adders, designed at three different abstraction levels: Gate Level, Dataflow Level, and Behavioral Level. These designs are fundamental to digital electronics, and this project showcases the versatility of Verilog in modeling and simulating digital circuits.

SUBIT‑64 is a universal semantic operating system built on six binary axes, an eight‑layer semantic stack, and a 64‑node hypercubic topology. It provides a minimal, interpretable, and dynamic framework for modeling meaning, cognition, behavior, and system transitions across domains.

Motivation tracker for ADHD builders. Uses pharmacokinetic decay curves to predict project abandonment. Because motivation isn't infinite - it's exponential.

A sophisticated Python framework for modeling and simulating psychological personality systems. This project enables dynamic personality evolution through event simulation, allowing for realistic modeling of how life experiences shape personality traits and psychological attributes.

The second version (V2) of Project Adam: A multi-agent framework using a hybrid Director-Actor architecture to synthesize group and private chat datasets with advanced MBTI profiling and dynamic social conflict matrices.

An end-to-end customer segmentation and behavioral modeling project. This repository showcases the development of an RFM model from scratch using SQL to classify customers and empower targeted marketing strategies.

Vivado VHDL

AI-powered social media simulation framework using LLMs for modeling Twitter dynamics during crisis events. Built for SBP-BRiMS 2024 Twitter Challenge.

Signal-based journaling that evolves into behavioral infrastructure. Deterministic-first, privacy-first, no engagement engineering.

Комплексная система для генерации реалистичных траекторий перемещения людей в городской среде. Проект сочетает паттерны проектирования, пространственный анализ с OSMnx и статистическую валидацию против реальных данных

Framework-agnostic behavioral signal engine in TypeScript. Rolling-window pattern detection, cooldown enforcement, and detector abstraction.

Dietary Destabilization Triangle assessment tool

Research repository focused on longitudinal behavioral governance, runtime accountability, and AI-human interaction systems.

In CSE332L: Computer Organization and Architecture Lab, I completed an assignment involving several Verilog tasks, where I solved the problems using behavioral, dataflow, and gate-level modeling.