hi, i am
suman
kunwar.

researcher and software engineer

download resume

about me

Researcher | Author | Creator of Deep Waste: Where Waste Meets Innovation

An experienced software engineer, researcher and author with two products that made it into the top 10 ‘Product of the Day' in Product Hunt, People Choice Award Recipient of Code for Humanity Open Hack Stockholm (2020), and Third Place Recipient (Asia Pacific) of 2019 Developer Circles from Facebook.

phone

936-465-8926

email

info@sumankunwar.com.np

website

sumankunwar.com.np

education

2018 - 2022

PhD in computer science

Selinus University of Sciences & Literature

Ragusa, Italy

Thesis topic "An effective Convolutional Neural Network (CNN) to reduce construction loss in JPEG compression" presented in AIGC2023.

2020 - 2021

PGCert in Entrepreneurship

Coventry University

Priory St, Coventry CV1 5FB, United Kingdom

Specializing in Strategic Change and Innovation and Creating Entrepreneurial Impact

2020 - 2021

GCert in Entrepreneurship

Deakin University

Geelong, Australia

Specializing in Exploring Socio-Economic Innovation and Critiquing Entrepreneurship



2014 - 2016

master of science in information technology

Sikkim Manipal University

Sikkim, India

Specialization on Open Source System and Cloud Computing

2012 - 2014

bachelor of science in information technology

Sikkim Manipal University

Sikkim, India

Specialization on Human Computer Interface and Artificial Intelligence

skills

Web development

90%

App development

85%

Research

97%

Deep Learning

90%

Database

90%

Analytics

85%

communication

97%

creativity

90%

experience

  • Jan 2023 - Present

    Creator/ Machine Learning Researcher

    D.Waste

    Houston, USA

    Developed a state-of-the-art waste classification machine learning model and successfully implemented it in an app designed to manage waste more efficiently. Ensured the app functioned seamlessly both online and offline and made it available on both Android and iOS platforms. Additionally, published several papers that benchmarked various deep learning models, explored green models, and created a custom waste dataset along with an image annotation tool.

    Jan 2023 - Present

    Creator/ Machine Learning Researcher

  • Jan 2022 - Oct 2023

    Frontend Consultant

    Latana GmbH

    Berlin, Germany (Remote)

    Assist in making decisions regarding architecture, scale, caching, security, hosting, and configuration concerning Mobile Surveys written in PreactJS with RecoilJS. Collaborate closely with peers, UX/UI designers, and product managers to establish features and project estimates, while managing existing codebases to maintain code quality, diagnose bugs, and incorporate new features. Organize regular, structured meetings to guide mentees and provide direction for their work and future careers.

  • Jan 2019 - Dec 2021

    Frontend Developer

    Dalia Research GmbH

    Berlin, Germany

    Partner with UX/UI designers and backend developers to transform design concepts into a fully functional admin dashboard using ReactJS. Craft clean, maintainable, and efficient code with ReactJS and Styled Components to develop a platform enabling people to participate in surveys. Enhance the speed and performance of the VueJS survey interface while collaborating with the Quality Assurance team to establish automated end-to-end testing procedures.

    Jan 2019 - Dec 2021

    Frontend Developer

  • Feb 2015 — Dec 2018

    Senior / Software Engineer

    Fusemachines Inc / Fusemachines Nepal

    Kathmandu, Nepal

    Lead the team in developing the frontend of the learning platform, written in Typescript and ReactJS, utilizing REST API. Be an integral part of the development team, contributing to the migration of the lead generation and campaign outreach platform from EmberJS to ReactJS. Collaborate with Backend Engineers, product managers, designers, and QA to design and structure the learning platform.

profiles

github
github
Medium
Medium
Kaggle
Kaggle
Product Hunt
Product Hunt
Google Scholar
Google Scholar
stackoverflow
stackoverflow

portfolio

publications

Convolutional Neural Network (CNN) to Reduce Construction Loss in JPEG Compression Caused by Discrete Fourier Transform (DFT)

AIGC Conference: 10.1007/978-981-99-7587-7_25

MWaste: An app that uses deep learning to manage household waste

 

CTE Journal: 10.3934/ctr.2023008

Strategies in JPEG compression using Convolutional Neural Network (CNN)

 

IJRPR Journal: 10.55248/gengpi.4.723.48514

Mapping of Land Use and Land Cover (LULC) using EuroSAT and Transfer Learning

RIG Journal: 10.32604/RIG.2023.047627

Managing Household Waste through Transfer Learning

IDWM Journal: 10.53623/idwm.v4i1.408

Malaria detection using image processing and machine learning

DOI: 10.48550/ARXIV.1801.10031

Comprehensive Image Quality Assessment (IQA) of JPEG, WebP, HEIF and AVIF Formats

DOI: 10.31219/osf.io/ud7w4

Deep Learning in Waste Management: A Brief Survey

IJCAST Journal: pending

Annotate-Lab: Simplifying Image Annotation

JOSS Journal: pending

Past clients

contact me

suman kunwar

Researcher / Software Engineer

phone

936-465-8926

email

info@sumankunwar.com.np

website

sumankunwar.com.np