Anurag Kumar

Anurag Kumar

Software Engineer

Contact Me

About Me

A versatile engineering student with a commitment to and experience of developing innovative and creative software solutions. Actively looking for a software engineering position with a fast-paced and ambitious company that can offer development training, project based learning and a mentoring programme.

My most recent interest is Machine Learning concepts. I had shifted from JAVA to python because ML needed me to. In my specialization course I had studied and implemented many ML concepts including regression, classification, clustering and retrieval. I am familiar with few techniques needed for good ML models like cost estimations, L1 L2 penalties, Gradients and boosting and many more. I am mesmerized by the idea of a soulless machine giving me sensible suggestions( product recommendation engines ).

Work Experience

ML Research Intern - Indycium Tech. (2017)

Working on various problems some of which require ML concept approach and others just basic python development skills. I have given most of my time in optimization and efficient extraction of native and contextual features from certain inputs.

Assoc Software Dev Engg - Soincwall Inc. (2018)

Python Development and DEV OPs

Latest Projects

WordNet

Create a Simple network of words related to each other using Twitter Streaming API.

project name
New

This is a open module I have developed for the soul purpose of creating network of contextually relative words, basically a simple form of word-play written in core python. As of now, this project includes an twitter streaming api helping to generate proper corpus for the network.

It uses TF-IDF values for words in given corpus as feature for a strong relation between words. My future goals for this module is to integrate more / better features for relation generation and more sources of corpus like facebook and reddit.

I am also planning to use graph-based databases for future integrations due to their similarity with the data I am generating.

View Source
project name

Project-WhatRU (Android application Project)

An Android Environment based project that predicts the gender of the user by asking their height and weight. This was my first attempt towards smart applications that don’t just compute, but think or analize before responding. For this I used the simplest form of k-means clustering of data under supervised (labeled) learning.

Find out more

project name

Project-Tweety (Arduino Project)

This project was basically my humble attempt in college where I wished to give something back to my college. To create a unique model and hand-it-over to juniors. It is simply an Arduino Circuit on a plant with humidity and temperature sensors which is linked to a twitter account handle which basically tweets intentional phrases at peak changes in humidity or temperature throught the day. In short, a tweeter handle for your favourite plant at campus.

Find out more

Trainings and Certifications

Maestro in SDDM using Java - Kyrion Technologies Pvt. Ltd. ,New Delhi, India Training

In this training, I learned basic JAVA web based services and application development and it’s lifecycle. I was introduced to RESTfull service designs and helpful tools like Hibernate ORM & Spring frameworks. As a final project to this training I developed a lightweight social networking web application that revolved arund the concept of friends sharing fun activity they did and encouraging others to do the same.

View on GitHub

Cloud Computing and JAVA – KVCH IBM-CE, Noidas Training

In this training, I revised my core JAVA skills and learned cloud computing based application hosting and development with bluemix staging. My final project towards this training was a model for Digitized Sales System. In which, I implemented basic aesthetics of a e-commerce application including ordering (carting) products, selling products, reviewing and product browsing.

View on GitHub

Machine Learning Specialization – University of Washington (coursera.org) Certification

In this certicfication course, I went through set of 4 courses covering standard ML concepts and model techniques like regression, clasiffication, clustering & retrieval. Some additional techniques for better understading and implimentation of the same were also provided like cost estimation, penalties, precision-recall, Gradients etc.

View Certificate

My GitHub

Calendar:

Loading the data just for you.

Languages:

Feeds: