Assessment Reporting Tool shortly called as ART. This tool developed for the company who are into verbal reasoning test. This tool used to analyze the lot of input files and generate the calculated output. The main objective of this tool is accuracy of the calculated result. When they calculate the result for a single candidate it will take almost 20 to 30 minutes. Using this tool they can generate the result for 1000 candidates within a few seconds. I have used MS Access as the front end and also as the backend. The challenging part in this project is accuracy, speed, performance and merging the result to the word document. Some outputs are exported to word format. I have to make sure that it should work with all the office editions.

It is a process of parsing the consolidated 835 file that comes in HIPAA-format file which is a merged multiple individual 835 files sent on the same day. This consolidated 835 file is tied to the automated payment application for that day's deposit to the bank account to ensure that the fund's transfer and ERA match one for one. The application also verifies if the total amount of the consolidated 835 is always equal the amount sent to the depository bank account that day. This enables to receive, and track, payment for multiple claims easily and effectively. The file 835 contains information about the payee, the payer, the amount, and any identifying information of the payment. In addition, the 835 can authorize a payee to have a DFI take funds from the payer's account and transfer those funds to the payee's account. I also created the intranet site to view the processed data.

The client company is a pharmaceutical chain headquartered out of Southern California. In addition to pharmacies, they are involved in the sale of hospital supplies, a workman's compensation claim service, and long-term care facilities. Built applications for them to automate payment processing, collections, and the workman's compensation department.

PlanMyLeave is a cloud-based online leave management system designed for multi-location small and medium enterprises. This web-based absence management system has a simple and user-friendly interface that gives management and employees immediate and transparent access to leave entitlements.

The existing product technology was built on opensource telephony software and the client was not satisfied with its performance. We designed and developed a new conferencing engine from scratch with performance as the focus. As a part of this built a memory efficient SIP and RTP protocol stack. The new system exceeded expectations and was capable of handling 8000 simultaneous phone calls on a mid-range desktop class machine. Each phone call involves transferring 100 packets per second, resulting in a total of 800,000 packets per second. We also built a clustering system, so that a cluster of conferencing servers behave as one large logical conferencing server. This clustering system has been tested to scale to 100,000 simultaneous phone calls.

Developed an intelligent web crawler that can accept a company's website URL, crawl the website, and extract the management team details. The software will enumerate the names and the titles of the team. This project involved Natural Language Processing, Probabilistic Modeling, and Heuristic Design.

Happyfox Chat is a web-based live-chat service for customer support. I have coded the entire backend for the chat system. It has 8 different servers each serves a different purpose. It is currently running across more than 100 servers due to the higher number of visitor traffic. Exported the client API library in different languages Swift - iOS, Objective C - iOS, Java - Android, Javascript - Web and NodeJS - Server. I mostly used python and used LMDB as a main database. Since LMDB is a key value database we created the custom data type system to work with LMDB.

This projects consists of below sub projects. I have designed, architected, and coded from scratch. Worked with the client directly.

- Admin Server (Python, LMDB)
- Coordinator (Python)
- Replication Server (Python, LMDB)
- Agent Availability Server (Python)
- Visitor Count Server (C++)
- Key Value Store (Python, LMDB)
- Company Server
- Front End Server
- Client API library for iOS, Android, Web and Server to Server. (Objective C, Swift, Java,Javascript and NodeJS)

Vaya Lynk is a bluetooth tracker. It is a small smart IoT device that you can attach to your personal belongings and connect to your phone using Bluetooth. It helps you track your things and locate them by ringing when they get misplaced or lost. You can get more information about the tracker from the shopping URL (https://vaya.in/smart-tracker)

This project consists of 4 sub projects. I have designed, architected, and coded from scratch. Worked the client directly

- IOS Mobile App (Swift)
- Android Mobile App (Kotlin)
- Application Server (NodeJS, Redis as a main database) - User Website (Nodejs, Angular JS)

It is a process of parsing the consolidated 835 file that comes in HIPAA-format file which is amerged multiple individual 835 files sent on the same day. This consolidated 835 file is tied tothe automated payment application for that days deposit to the bank account to ensurethat the funds transfer and ERA match one for one. The application also verifies if the totalamount of the consolidated 835 is always equal the amount sent to the depository bankaccount that day. This enables to receive, and track, payment for multiple claims easily andeffectively. The file 835 contains information about the payee, the payer, the amount, andany identifying information of the payment. In addition, the 835 can authorize a payee tohave a DFI take funds from the payers account and transfer those funds to the payeesaccount. I also created the intranet site to view the processed data.

SoftwareAssessment Reporting Tool shortly called as ART. This tool developed for the company whoare into verbal reasoning test. This tool used to analyze the lot of input files and generate thecalculated output. The main objective of this tool is accuracy of the calculated result. Whenthey calculate the result for a single candidate it will take almost 20 to 30 minutes. Using thistool they can generate the result for 1000 candidates within a few seconds. I have used MSAccess as the front end and also as the backend. The challenging part in this project isaccuracy, speed, performance and merging the result to the word document. Someoutputs are exported to word format. I have to make sure that it should work with all theoffice editions.