Building ProfileFusionAI

github

Davnish Singh  • Peerlist

ProfileFusion is an innovative platform designed to help you create stunning professional profiles that truly stand out in the digital world. With AI-powered endorsements, you can showcase your skills in a way that is both impactful and credible. This tool is perfect for professionals looking to enhance their online presence and make a lasting impression. Key features include the ability to build comprehensive profiles that highlight your skills and achievements, as well as leveraging AI technology to provide endorsements that add value and authenticity to your profile. Whether you're a job seeker, freelancer, or a professional looking to network, ProfileFusion offers the tools you need to create a compelling professional identity. Join the digital revolution with ProfileFusion and take control of your professional narrative.

ProfileFusion

•Designed and implemented console app in C++ for managing personal details of employees.
•It can add, remove and listing information of employees.

Banking Record System

• Bill splitting app splits the bills depends on the number of people contributing to pay.
• Front end designed in Visual Studio 2017.
• Backend developed and executed in C++ and C#.
• All data is stored in access DB created in Microsoft access with queries loading much efficiently.
• System is developed and debugged in visual studio with all the test cases passed in one-go.

Bill Splitting App

• Developed the entire platform and implemented new algorithm CW Rank which ranks all the
   users and their confessions based on the total views, CW Rank of a user and performance factor.
• Implemented and created the entire user-based analytical system where users can analyze and
   view the current analytics of their profile.
• Stored user data in MySQL database with clustering and sharding have done to store and retrieve data 
   more efficiently. 
• Increased user base on 12% monthly with 5.2k views monthly on an average user’s combine
  confessions.

ConfessionWeb

•Data of clients handled with TopProducer 8i classified with name, phone number and contact.
•Developed a python script to scrape data from website
•Stored and read the data from SQL
•Took all the data from database and entered into the excel spreadsheet with xlwings library.
•Scraped around 2.5k queries within 48seconds.

Customized Advanced Data Scraper

•Designed and Implemented this app with HTML, CSS and JavaScript.
•Developed this with the help of weather API which automatically updates the weather information according to your current location. 
•Parameters of API respond Min/Max temperature in current weather API and forecast API.

Local Weather App

•Designed and implemented a HTML, CSS and Bootstrap based website.
•Implemented a sorting algorithm to sort all the trailers in a list based on their releasing date and views.
•Parsed Movie trailers from youtube channel with Youtube Data API v3 and jQuery.
•Implemented a method in which Youtube API will automatically post the latest video to website and add to the list of trailers based on their views and time.

MovTrailers - jQuery based movie trailers rec

•Designed and implemented console app in C++ for managing personal budget.  
•Provided interactive internal command line to add, remove and list purchases

myBudget

•Developed and deployed my own website.
•Implemented this project in HTML, CSS(Sass), Bootstrap and Javascript for front end Development.

Personal Website

•Developed this leaderboard using HTML, CSS and JavaScript. 
•Implemented current stream feature with the stream API.
•This board will show current streak of a programmer for number of days.

Programmer Leaderboard

•Developed this game using HTML(Jade), CSS(SCSS) and JavaScript(Babel).
•It is characterized by a dungeon crawl through procedurally generated game levels, turn-based gameplay, tile-based graphics, and permanent death of the player-character.

Roguelike Dungeon Crawler Game

Developed and implemented the game using HTML, CSS and JavaScript.

Simon Game

TaskHive is a decentralized task outsourcing platform that connects users with local freelancers through a competitive bidding system. Designed to simplify small-job hiring, the platform enables real-time task posting, bidding, communication, and verification.

Key Contributions:
	•	🛠 Built the MVP from scratch using the MERN stack with full ownership of both frontend and backend.
	•	⚙️ Designed and implemented a dynamic task bidding system where users can post jobs and freelancers place bids in real time.
	•	🔐 Integrated secure user authentication using JWT and role-based access for users and taskers.
	•	📦 Developed a detailed task workflow system including task status transitions, bid acceptance, and progress tracking.
	•	📊 Implemented a task verification & approval module allowing clients to verify and close completed tasks.
	•	💬 Built interactive UI components with React and Tailwind CSS for a smooth, responsive user experience.
	•	🧪 Wrote unit tests for core backend APIs and ensured seamless integration between client and server.
	•	🚀 Deployed MVP early for testing, with future plans to integrate payment gateways (Stripe/PayPal) and geolocation-based task filtering.

TaskHive

• Authorized Twitter API client.
• Made a GET request to Twitter API to fetch tweets for a particular query.
• Parsed the tweets. Classified each tweet as positive, negative or neutral.
• Used sentiment.polarity method of TextBlob class to get the polarity of tweet between -1 to 1.
• Resulted with Positive Tweets: 25.31% and Negative Tweets: 13.92%.

Twitter Sentiment Analysis (Python)

1) Implemented the backend using PHP and Frontend using HTML, CSS(Bootstrap).
2) Data storage is handled using MySQL DB with dependencies and references checked for all of the tables.
3) Implemented the search feature for the users to search for cars using Car name, Car Model and the Location.
4) Implemented and developed the Admin platform to handle the user data, cars, reservations, rentals and producing bills.
5) Developed the User platform to let the users find the cars, reserve them and rental the cars once be confirmed by the admin of the platform.
6) The platform is developed to handle 250+ concurrent users tested using www.impactload.com