Engineering & Dark Arts | Building Plum

devto

github

substack

medium

productHunt

Harsh Joshi  • Peerlist

This project aims to work on developing AI models based on Generative Adversarial Network(GANs) to draw art portraits using GAN models and work on efficiency using mathematical and statistical measures like the golden ratio. 
The main objective of the project is to develop and model the Generative Adversarial Network (GANs) and train them to draw art portraits. The project also aims to work on the efficiency of Discriminator learning to produce better results.

AI Art using Generative Adversarial Networks

Here are the projects:

1-QBox
QBox is a secure file-sharing service. It is a customized version of the popularly available VSFTPD server.

-- Created pre-install scripts to run system checks, install and configure QBox.

-- Developed tools to automatically analyze performance and functionality issues in different versions of QBox.

2 - QMoney
QMoney is a visual stock portfolio analyzer. It helps portfolio managers make trade recommendations for their clients. 

-- Implemented the core logic of the portfolio manager and published it as a library.

-- Refactored code to add support for multiple stock quote services.

-- Improved application stability and performance.

3 - QPrep
Learned and Practiced algorithmic problems in a structured way.

-- First articulating the problem statement clearly with examples
-- Brainstorming multiple approaches, understanding trade-offs and writing pseudocode for the chosen approach -
-- Implementing the solution and testing it

Crio Launch Projects

This project aims to compare and summarize the computer network routing strategies while investigating the optimization of these using PSO technique. The project focuses on minimal spanning tree and solves the shortest path problem using swarm intelligence. Particle Swarm Intelligence utilizes the combination of the behaviour of swarms which is considered efficient for optimization and discrete multidimensional problems. In Particle Swarm Optimization Algorithm, a subset of Swarm Intelligence, we optimize network routing by modifying traditional graph algorithms like Minimum Spanning trees by Kruskal’s or Prim’s methods by using cognitive intelligence algorithm to get better optimization and solve MDR(Multi-Destination Routing) problem which cannot be solved effectively using traditional methods.
Particle Swarm Optimization algorithm implementation on network routing gives fast convergence speed, has an easy implementation, is cognitive and supports dynamic traffic.

Network Route Optimization using Particle Swarm Optimization

Calo is an intelligent healthy utility android application. It is AI-powered and utilizes cloud services to be lightweight, efficient and secure. 

Core Functionalities:
1 -Calorie Detection in Unpacked Food just by clicking a picture of it using Object recognition algorithm with accuracy ranging from 90-100%
2 -Monitoring Heart Rate (Pulse Rate) without needing any extra sensor (Based on Monochromatic Red Light sensing using LED)
3 -Monitoring Blood Pressure
4- Tracking the physical activity of a user based on step count detection and body movement.
5 -Intelligent Medicine Information and Drug Detection using Optical Character Recognition and Drug APIs
6 - Robust Security Mechanism implemented using Google firebase Authentication
7 - Removes Fraud and Spam User threat by two-way email verification.
8 - Scalable, Multiple android API supporting user interface with Material UI implementation. (Works for 100% android devices > API 21)
9 - Habit Tracker which helps to set goals to monitor the progress resulting in strong will towards goals affecting the emotional well being.
10 - Complete and Thorough Health Analysis which as large as 30 days analysis on nutrition intake, Heart beat Trends, Mood Trends and lots more.