Arpit Roopchandani

Full-Stack Engineer & AI Developer

mail@arpit.io github.com/whoisarpit linkedin.com/in/whoisarpit
Summary

Engineer with 7+ years of experience in DevOps, and generative AI technologies. Currently building an AI-powered code fixing platform at Patched, backed by Y Combinator. Passionate about creating innovative solutions that transform how developers work.

Skills
React TypeScript NextJS Tailwind Generative AI LLMs Python Docker AWS CI/CD DevOps
Work Experience
Patched (YC S24) Sept '23 - Present
Founding Engineer

Developing an AI-powered code fixing platform, building full-stack infrastructure using TypeScript, NextJS, and generative AI technologies.

Bunker Technologies Feb '23 - Aug '23
Individual Contributor - Smart Reports

Led the ideation and development of Smart Reports, the first data-driven insights feature for the flagship product, providing actionable insights from past financial data to improve strategic decision-making.

Cyber Security Research Centre @ NTU Jan '19 - Sept '23
Research Engineer

Led the Frontend and Devops development and made contributions to the backend for a comprehensive home care management and allocation platform, leveraging data insights.

qplum Dec '17 - Dec '18
Frontend Developer

Developed the website and mobile app for the Quant-based Robo-advisory firm, offering AI and machine learning based portfolios. Led the switch to goal-based investing.

Finvolv.ai Feb '17 - Nov '17
Frontend Developer

Developed the website for one of India's largest asset management companies, UTI Mutual Funds. Developed the banking website for HDFC, one of India's largest private banks.

Education
M. Eng. - Computer Science (Part Time) Jan '21 - Jan '24
Nanyang Technological University, Singapore
Thesis: Preference-Aware allocation and scheduling of teaching duties and through multi-constrained optimization
B. Tech. - Computer Science Aug '13 - May '17
Jaypee University of Information Technology, India
Projects/Contributions

Open source projects

  • form-it - Drag and drop form builder for digitizing internal forms.
  • vue-modal-promise - Promise-based modal library for Vue.
  • js-pick-file - Multi-platform library to open a promise-based file picker.
  • vue-resource-timeline - Component for viewing events on a calendar for multiple resources.

Editors Choice Classification for News Comments

Using Deep Learning and NLP to predict if a comment will be selected as Editors Choice.

Image Captioning for COCO Dataset (2021)

Comparative study of image captioning models on the COCO Dataset.

Sofar Sounds - City Co-Lead and Artist Curator

Organized intimate house-concerts, managing logistics, sound, and sponsorships.