Aditi Swaroop
Fullstack Software Engineer | GyShiDo-San
About
Hello and Welcome!
I have been working as Software Engineer for a startup, Jobox.ai in Palo Alto.
Jobox.ai is a marketplace for home service technicians. I was an individual contributor in driving the core dispatch flow that sends jobs to these techs along with working on several other high-impact projects in the company.
Read about my work at Jobox here.I graduated from the University of Southern California with a Master's in Computer Science, in December 2020.
At USC, I worked as a web developer for a Human-AI Interaction research project. I also published my first iOS app during an internship at a seed stage startup, while pursuing my degree at USC.
Before moving to Los Angeles, I was working with Cognizant in India, where I gained three years of significant work experience being a part of a multinational cross-functional development team.I turn coffee into code, deadlines into deliveries,
and problems into software solutions.
Not an "all-knower" but definitely an "all-learner". Learning is a continuous process and I am enthusiastically grasping new programming languages, frameworks, and principles as I move forward in my career.
Education
| University of Southern California, Los Angeles | M.S in Computer Science | GPA 3.7 | Course Producer, CSCI-420 Computer Graphics |
| SRM University, India | B.Tech in Computer Science | GPA 8.95 | Grader, Analysis of Algorithms | Basketball team captain |
Skills
| Programming Languages | Java | Python | C++ | C# | Swift | |||
| Web Technologies | VueJs | ReactJs | NodeJs | Javascript | Typescript | HMTL+CSS | Bootstrap | jQuery |
| Databases | SQL | PostgreSQL | SQL Server | MongoDB | ||||
| Frameworks and Tools | REST APIs | Hibernate/Spring | Jenkins | AWS | Kibana | ARKit/SceneKit | ASP.NET | Gradle |
| Prototyping and UML | Figma | AdobeXD |
WORK EXPERIENCE
Sofware Engineer at Jobox.ai, Palo Alto (Feb 2021 - Feb 2023)
Jobox connects companies with thousands of vetted home service professionals through a marketplace built on an AI-based infrastructure as a one-point solution for jobs, schedules, and expense management.
• Subject matter expert for the core dispatching algorithm responsible for matching a job to skilled technicians based on several internal parameters. Reduced time for matching jobs to qualified technicians from 10 minutes to <20 seconds by automating the flow.
• Individually built features like multiple dispatches and automated re-dispatches on top of the existing algorithm, thus increasing the throughput of job acceptance by 30%, making a high impact on the company's revenue
• Opportunistically led and developed the job lead management infrastructure, enabling analytical insights, data-driven decision-making, and creating a ramp for introducing ML in the system
• Worked closely with CTO and product team, contributing to technical discussions in brainstorming product enhancements, reducing tech debts, and creating processes for seamless deployments of servers.Recognized for engineering contributions and building company culture. Read my article here.
Software Engineer at Opal AI, Los Angeles (June 2020 - Feb 2021)
OPAL AI utilizes LiDAR and AR technology on mobile phones to streamline floor-plan creation and eliminate time-consuming manual overheads like using a tape measure, repeated site visits, and scope for errors etc.
• Picked up Swift and ARKit knowledge piloting end-to-end development and launch of the MVP mobile app - ScanTo3D - thus facilitating the foundation for seed funding
• Expedited floor-plan generation by 90% through automated generation using point cloud data gathered from augmented meshes. Set up the pipeline for data collection and storage for running ML algorithms and improving the floor plan creation
• Researched solutions and developed a web platform tied up with the app to allow users to view the 3D model of their captures, instantly, with features like taking and recording measurements mapped to real-life metrics.
Software Engineer at Cognizant, India (September 2015-November 2018)
Collaborated with cross-functional teams working for a health insurance client – Wellmark, Inc. on a workflow migration from a legacy system to Cognizant TriZetto’s Facets® - a healthcare administrative platform
• Wrote technical and functional design documents adhering to the process and coded C# extensions tailored to business requirements on the existing Facets tool.
• Performed impact analysis, defect detection, and setting up unit test cases to ensure software quality.
• Analysed and rewrote existing functions and SQL procedures to improve underperforming processes, achieving a run time improvement of up to 40%
• Outside of work duties, took the initiative to build an internal web-based resource management system from scratch. This helped the leadership team track project metrics like resource allocation, budget management, visual representation of project progress, etc, and eliminated the need to maintain data in Excel sheets.
• Surpassed entry-level trainees in skill assessments to stand amongst top 50 candidates nationwide selected for self-learning based ‘Fastrack Training’
Work Experience at University of Southern California
Web Developer, USC Annenberg (May 2019-April 2020)
• Materialized a web-based game to support a research study at USC Annenberg School of Communication backed by Microsoft Research.
• The study was based on human-AI interaction and how performance, human interaction, trust, information exchange, and ethical behaviors are influenced under competitive and pressured environments.
• Created a conversational bot to explore how AI assistants might serve as an intervention to nudge decision-makers to increase information sharing.
Projects
corona token
This is a concept project - a web app designed to support social distancing during Covid-19.
Created on ReactJS, the idea is to prevent overcrowding public places like beaches, museums, parks, etc. in LA by allocating a fixed number of tokens for visiting a location.
The number of tokens is calculated daily based on the Covid-19 curve around that location. Users may visit a place only if they have a token, thus ensuring only a certain number of people visit the beach in a day!
Iron Man on Moon
A POC application I built to conceptualize Swift and ARKit learning.
This app works on a ray-casting algorithm. A "hit test" is performed to determine to world coordinates of objects and using ARKit4 move the object, in this case, Iron Man to the determined locations.
Weather Forecast App
A mobile and web app that lets users enter a location and retrieve its weather information for the upcoming 10 days.
By implementing Google's geolocation API to collect location information and Dark Sky Weather API, weather information was attained in JSON format. The application also utilizes local storage to store users' favorite locations.
Open Sesame
A web game experiment to determine patterns of information exchange between humans in permutations of competitive environments.
An interactive bot - Alana - was created to study if AI conversational bots play a role in critical decision-making in such situations
The study was further extended to analyze if the gender of these bots has an effect on the level of trust and reliability of users.
Simulating a roller coaster
The project uses Catmull-Rom splines along with OpenGL core profile shader-based lighting and texture mapping to create a roller coaster simulation.
The simulation runs in a first-person view, allowing the user to "ride" the coaster in an immersive environment.
Winbook 2.0 (WinSC)
This is a requirement gathering and negotiation web tool currently in production at USC CSSE and the software engineering class.
The original application, which was coded in Java, was revamped from scratch using Angular and NodeJS to create a more modern, responsive tool.