Hey! I'm Jainil.
I
for fun.

Hey!

My name is Jainil, and I recently graduated from UC Berkeley with a degree in computer science.

As a child, I would often find myself solving riddles and fiddling with the classic Windows 7 sliding puzzle. Although I parted ways with my antiquated PC in 2013, my love for puzzles remained intact. Programming was a natural extension of this passion; I relish the challenge of solving "algorithmic" puzzles (except with real impact now!). On this website, I've attempted to catalogue some of the puzzles I've tinkered with and experience I've accrued.

This past year, I taught database concepts and helped build a novel database for managing multimodal log data. Presently, I'm headed to Plaid to explore problems at the intersection of software, systems, and security.

In my free time, I enjoy playing cricket, climbing (rocks, trees, walls, you name it), dancing, and geeking out over sports.

Work

Software Engineer Intern, Security // Plaid

May 2023 - Aug 2023
Led an organization-wide effort to revamp AWS Identity and Access Management (IAM) at Plaid.
Built a new permissions management system in Go to support fine-grained, time-bound permissions for AWS resources.
Implemented Terraform auto-generation, slashing latency in requesting new AWS IAM roles from 2 weeks to 3 days.
Pioneered a template to dynamically generate IAM policies at runtime and reduce policy bloat by ~90%.
Exposed audit logs for employee permissions, highlighting permissions for removal to ensure least-privileged access.

Undergraduate Research Assistant // Berkeley Artificial Intelligence Research

Aug 2022 - Aug 2023
Worked on the state of the art in learning-based cable tracing and untangling. Advised by Prof. Ken Goldberg.
Devised and implemented a novel crossing correction and knot detection algorithm for a robotic cable untangling system.
Published with an oral presentation (6.6% of all submissions) at the Conference on Robot Learning (CoRL) 2023.

Software Engineer Intern // Salesforce

May 2022 - Aug 2022
Built a first-of-its-kind search feature for Composer, enabling users to discover their previously inaccessible data lineage.
Won the Best Customer Value Hack at an organization-wide hackathon and led a live customer demo.
Developed and shipped a user-requested feature to support custom logic in MuleSoft Composer's condition cards.
Used React, Redux, and TypeScript on the frontend. Worked with Java (Spring) microservices on the backend.
Wrote tests (in JUnit, Jest, and Mockito), fixed bugs, and handled feature planning alongside the team's PM.

Co-Founder // Track My Change

May 2021 - Dec 2021
Founded a 6-person startup to quantify the impact of social organizations. Launched a pilot for customers in the US and Nepal.
Created customizable web widgets to convey social impact to convey social impact to potential volunteers and donors.
Created REST API in PHP with MySQL to track analytics. Monitored widget interactions with JavaScript.

Software Engineer Intern // Jacobs

Jun 2021 - Aug 2021
Developed a sentiment classifier in Python using a RoBERTa-based model to assess customer perception of Jacobs.
Queried post data using the Facebook Graph API & LinkedIn API. Displayed insights on a Power BI dashboard.

Software Engineer Intern, Security // Plaid

Aug 2022 - Present

Led an organization-wide effort to revamp AWS Identity and Access Management (IAM) at Plaid.
Built a new permissions management system in Go to support fine-grained, time-bound permissions for AWS resources.
Implemented Terraform auto-generation, slashing latency in requesting new AWS IAM roles from 2 weeks to 3 days.
Pioneered a template to dynamically generate IAM policies at runtime and reduce policy bloat by ~90%.
Exposed audit logs for employee permissions, highlighting permissions for removal to ensure least-privileged access.

Undergraduate Research Assistant // Berkeley Artificial Intelligence Research

Aug 2022 - Present

Worked on the state of the art in learning-based cable tracing and untangling. Advised by Prof. Ken Goldberg.
Devised and implemented a novel crossing correction and knot detection algorithm for a robotic cable untangling system.
Published with an oral presentation (6.6% of all submissions) at the Conference on Robot Learning (CoRL) 2023.

Software Engineer Intern // Salesforce

May 2022 - Aug 2022

Built a first-of-its-kind search feature for Composer, enabling users to discover their previously inaccessible data lineage.
Won the Best Customer Value Hack at an organization-wide hackathon and led a live customer demo.
Developed and shipped a user-requested feature to support custom logic in MuleSoft Composer's condition cards.
Used React, Redux, and TypeScript on the frontend. Worked with Java (Spring) microservices on the backend.
Wrote tests (in JUnit, Jest, and Mockito), fixed bugs, and handled feature planning alongside the team's PM.

Co-Founder // Track My Change

May 2021 - Dec 2021

Founded a 6-person startup to quantify the impact of social organizations. Launched a pilot for customers in the US and Nepal.
Created customizable web widgets to convey social impact to convey social impact to potential volunteers and donors.
Created REST API in PHP with MySQL to track analytics. Monitored widget interactions with JavaScript.

Software Engineer Intern // Jacobs

Jun 2021 - Aug 2021

Developed a sentiment classifier in Python using a RoBERTa-based model to assess customer perception of Jacobs.
Queried post data using the Facebook Graph API & LinkedIn API. Displayed insights on a Power BI dashboard.

Software Engineer Intern

Plaid

Led an organization-wide effort to revamp AWS Identity and Access Management (IAM) at Plaid.
Built a new permissions management system in Go to support fine-grained, time-bound permissions for AWS resources.
Implemented Terraform auto-generation, slashing latency in requesting new AWS IAM roles from 2 weeks to 3 days.
Pioneered a template to dynamically generate IAM policies at runtime and reduce policy bloat by ~90%.
Exposed audit logs for employee permissions, highlighting permissions for removal to ensure least-privileged access.

Undergraduate Research Assistant

Berkeley Artificial Intelligence Research

Worked on the state of the art in learning-based cable tracing and untangling. Advised by Prof. Ken Goldberg.
Devised and implemented a novel crossing correction and knot detection algorithm for a robotic cable untangling system.
Published with an oral presentation (6.6% of all submissions) at the Conference on Robot Learning (CoRL) 2023.

Software Engineer Intern

Salesforce

Built a first-of-its-kind search feature for Composer, enabling users to discover their previously inaccessible data lineage.
Won the Best Customer Value Hack at an organization-wide hackathon and led a live customer demo.
Developed and shipped a user-requested feature to support custom logic in MuleSoft Composer's condition cards.
Used React, Redux, and TypeScript on the frontend. Worked with Java (Spring) microservices on the backend.
Wrote tests (in JUnit, Jest, and Mockito), fixed bugs, and handled feature planning alongside the team's PM.

Co-Founder

Track My Change

Founded a 6-person startup to quantify the impact of social organizations. Launched a pilot for customers in the US and Nepal.
Created customizable web widgets to convey social impact to convey social impact to potential volunteers and donors.
Created REST API in PHP with MySQL to track analytics. Monitored widget interactions with JavaScript.

Software Engineer Intern

Jacobs

Developed a sentiment classifier in Python using a RoBERTa-based model to assess customer perception of Jacobs.
Queried post data using the Facebook Graph API & LinkedIn API. Displayed insights on a Power BI dashboard.

Projects

dp3

I co-developed dp3, a database with a novel approach to managing multimodal log data, such as logs produced from robotic sensors. I added support for range deletion of messages, logical data truncation, and reverse time-ordered traversal of messages.

dp3

Co-developed dp3, a database with a novel approach to the management of multimodal log data.

Secure File Sharing System

I developed an end-to-end encrypted file management system in Go. The app featured secure support for user authentication, file storage and retrieval, and file sharing and revocation. My implementation ranked 1st in a class of 650 students.

Secure FIle Sharing System

Java, Git

Built an end-to-end secure application that allows users to store and share files (ranked 1st out of 650 students).

Version Control System

Java, Git

I built a miniaturized version control system in Java. The system leveraged file persistence and cryptographic hashing. I added support for remote functionality alongside core version control features (e.g. commit, checkout, merge).

Version Control System

Java, Git

Built a miniaturized version control system. Implemented remote functionality alongside core VCS features.

Thought Processors

Django, PostgreSQL, JS

I co-founded an online platform for hosting cryptic hunts. Over the course of 3 years, we organized 4 events and accumulated a user base of 250 students (with 10,000 answer submissions). The platform was written in Python and built using Django and PostgreSQL.

Thought Processors

Django, PostgreSQL, JS

Co-founded a platform to host India's first quest-based cryptic hunt (250 users with 10,000 submissions).

Food Delivery Web App

Flask, PostgreSQL

I launched a mobile-responsive, full-stack web app to display trending food items. Food orders were evaluated in real time against custom-built heuristics. The web app was built with Flask and PostgreSQL and deployed with Heroku. Materialize CSS was used for styling.

Food Delivery Web App

Flask, PostgreSQL

Launched a full-stack web app to evaluate food orders against custom heuristics and display trending orders.

Heart Disease Classifier

Python

I developed a classifier using Principal Component Analysis (PCA) and Support Vector Machines (SVM), achieving an accuracy of over 95%. I analyzed multiple cardiovascular disease datasets and visualized trends with Seaborn and Plotly.

Heart Disease Classifier

Python

Developed a PCA and SVM classifier with an accuracy of over 95%. Visualized trends with Seaborn and Plotly.

Hey! I'm Jainil. I for fun.

Hey! waving_hand