A production-grade network intrusion detection system combining signature-based detection, anomaly detection, deep learning, and ensemble methods. Processes PCAP data in real-time to detect, classify, and analyze cyber attacks.

A fine-tuned YOLOv8 Nano-based face detection system optimized for detecting small, distant, and low-resolution faces in real-world CCTV and crowd surveillance environments.

An AI-powered VS Code extension that autonomously analyzes, restructures, and documents Python projects using Google Gemini — transforming legacy codebases into clean, maintainable architectures.

A farm-level crop recommendation engine built using satellite-derived soil, weather, and geospatial data. Analyzes environmental conditions to suggest optimal crops and improve agricultural decision-making.

An AI-powered newspaper processing system that extracts, structures, and enhances textual content from scanned images using column-aware OCR, computer vision, and language models. Converts cleaned text to audio for accessibility.

An AI-accelerated pipeline for generating high-resolution Lunar Digital Elevation Models from mono imagery using physics-based Shape-from-Shading combined with deep learning. Designed for mission-grade rover navigation and hazard mapping.

A production-grade AI-powered career intelligence platform combining psychometric analysis, real-time market data, and multi-agent AI coaching to generate personalized career roadmaps and simulate interviews at scale.

Secure load-forecasting pipeline using CKKS homomorphic encryption and differential privacy, enabling encrypted time-series inference without exposing plaintext energy data.

Autonomous rover with voice, mobile, and signal control modes. Includes face recognition, object tracking, line following, and obstacle avoidance using embedded computer vision. Built for ₹12K.