Abhishek Soni

● Design a multi platform dashboard to handle all your business request from one place and facilitate a customer engagement program for the customer ● Experience of handling 20 million users Product with daily request of around 250 million - 300 million. ● Worked on Profoundly Facebook game(backend) and handle a system which have 20M user base and almost 200M+ Daily message transfer between users (https://www.facebook.com/ProfoundlyBot/) ● Worked on Easemytrip Quiz system backend and create a serverless architechture to handle to quiz using ​ elastic beanstalk, Dynamodb, Aws Lambda, cloudwatch events to trigger Quiz and also create a CI/CD pipeline for auto deployment ● Worked on ​ AWS ECS ​ service using Fargate and create a plug and play micro application which is used to recommend users to chat in the system based on their preferences and also create its CI/CD pipepline for auto deployment ● Worked on audio Call system for our profoundly application and create a socket protocol to make sure client is connected to the system while making p2p audio calling (​https://profoundly.me/ ) ● Worked on Customer Support panel for ​ https://www.easemytrip.com/​ which is connected to their whatsapp number user send message to the number it goes direclty to the unassign panel and agents can pick chats and assign to ourself and talk with them there is also support for Sending Attachment to user and vice versa and use firestore db to make this chat realtime. ● worked on internal Blocking panel for Profoundly to view reported users activity and take actions based on that ● Worked on Static website and host it to ​ S3​ and create its CDN using ​ cloudfront​ to make it highly available to the end user(​https://neargroup.me/​ )

* XMPP(extensible messaging and presence protocol) Based Chat: 1- Multi user chat, One to one chat, Vcard Support, 2- Using Bosh over XMPP to make a bridge between HTTP and TCP. 3- XMPP node can handle 1, 50,000 active user on a single node * Multiclass image classification attendance system using Transfer learning and Keras. 1- Collected all the frontal face videos of all the classes and extracted snapshots of frontal faces using OpenCV. 2- Leveraged VGG16 pre trained weights architecture for transfer learning. 3- Implemented data augmentation and normalization techniques. 4- Trained a Convolutional Neural Network to classify images of all the employees to their respective classes. 5- Placed an automated attendance system using video surveillance camera. 6- Developed an internal feed and notification platform for Employees and HR using Django Framework. * Vehicle license number plate extraction system using OpenCV and Tesseract OCR. 1- Extracted vehicle image from live feed using object detection model. 2- Applied techniques for image preprocessing (Adaptive thresholding, Contour finding, Structured elements) to find possible plate in image. 3- Implemented character segmentation techniques to extract area of possible characters.

Doing freelancing in Blockchain and cryptocurrency develops robust application for my clients like:- * Crypto wallets integration, * Developing customize exchange application by merging two or more exchanges based on Django, * Develop arbitrage with 10 trading exchanges which support automatic fund redistribution when wallet funds go down to set threshold with the maximum amount wallet exchange * Develop a system for mining using antiminer s9 which can mine coins based on intelligent decision making system which tells which coin to mine to get more profit and also handle switching of miners over the coin * Design a platform which can fetch data from 60 exchanges with all supported pairs and store it in database to identify trends in the prices of coins

Experience & Responsibilities: Design & Developed robust web portals using Django/python, like Insurance sites, shopping sites, search engine, blog, social site. Develop android application for a health care Wearable device having Bluetooth low energy. Develop API using Django rest framework for a health care Wearable device Also, work with an embedded developer on the firmware of a health care Wearable device.(visit spiqi.com for more info) Learning : Lead a project from start and successfully complete it on time: * Design database schema. * Design Statement of Work (SOW). * Develop API using Django rest framework. * Also, work on an android application. Design silicone mold for the Wearable health care device cover and cast molded parts using hand molding technique

A humanoid robot that is capable of copying the physical motions of a human, pertaining to arm movements, leg movements and motions of all the major joints of the human body. Managing a team of 15 students, and worked on the Central Processing Unit of the robot ( Raspberry Pi), inclusive of developing and optimizing algorithms and modules.

A student startup, to work on innovations and converting ideas into real world technologies. As a self-motivated individual, worked on several projects, including Jugnu(Patented).