# Sujit Varadhan > Founding Engineer @ Hanover Park Location: New York, New York, United States Profile: https://flows.cv/sujitvaradhan Experience building AI agents and full-stack E2E features at Hanover Park, agentic AI workflows at Instalily, web accessibility tools at ClassTranscribe, and recommendation systems at Pandora. ## Work Experience ### Founding Software Engineer @ Hanover Park Jan 2024 – Present | New York, New York, United States ### Software Engineer @ InstaLILY AI Jan 2024 – Jan 2024 | New York, NY • Architected custom graph-based Python AI agent framework for multi-agent systems with configurable model switching, LLM tool calling abstraction, and execution tracing for agent workflow observability and debugging. • Created competitor analysis website used daily by 50+ marketing teams by building Python/FastAPI backend with Redis caching to serve ad copies/images, and React/JS dashboard with KPI cards, charts, and ad copy previews • Scraped 100+ competitor sites and Facebook Ad Libraries daily using automated Python/Selenium scraper with anti-bot bypass techniques including proxy rotation; deployed on GCE; stored ad text in Firestore and assets in GCS • Built sub-1 second latency AI voice agent for natural phone conversations with interruptable responses and noise-resistant voice activity detection, using Python, Twilio, OpenAI, Docker, and GCP Cloud Run ### Research Assistant @ ClassTranscribe Research Group Jan 2021 – Jan 2024 | Urbana, IL • Shipped 50+ features to 10,000 users within a production lecture video platform by taking leadership role in developing new assistive technologies to improve user experience for persons with disabilities, using React/Redux/JS. • Designed feature to create digital notes from lecture videos with real-time editing and previewing that mirrors final output across ePub, PDF, HTML, and LaTeX exports, using React, Redux, and JavaScript • Improved accessibility for blind students by creating audio description feature to generate, edit, and speak image descriptions within video player with adjustable text-to-speech parameters and screen reader compatibility ### Research Assistant: UDL and Accessibility Research Group @ University of Illinois Urbana-Champaign Jan 2021 – Jan 2023 • Co-authored five peer reviewed papers assessing the needs of undergraduate students in engineering classes and how to use new assistive technologies to improve education for students with disabilities. • Designed surveys and conducted interviews to assess the opinions and needs of students and instructors in undergraduate courses. • Designed training course to teach faculty best practices in making courses accessible for all students using Universal Design for Learning (UDL) principles. ### Course Assistant: CS 411: Database Systems @ University of Illinois Urbana-Champaign Jan 2021 – Jan 2021 • Assisted over 400 students with their group work during discussion sections. • Reviewed and wrote homework and exam questions. • Graded and provided meaningful feedback on students' code, homework, and exams. ### Research Assistant: Disruption Lab: Supply Chain Simulator Project @ University of Illinois Urbana-Champaign Jan 2020 – Jan 2020 • Utilized the Etherium Blockchain and wrote Solidity Smart Contracts to simulate an automated, traceable, and decentralized supply chain network. • Simulated the Supply Chain Pipeline by designing and implementing farmer, carrier, and manufacturer smart contracts. • Implemented automatic payments and refunds to ensure fairness across the supply chain. ### Software Engineer Intern @ Pandora Jan 2022 – Jan 2022 | Oakland, CA • Shipped 7 new algorithmic improvements to 40 million Pandora listeners within main song recommendation backend service, using Java, Spring Boot, and JUnit. • Enabled listening history A/B testing by refactoring Java/Spring microservices for concurrent reads/writes across multiple Bigtable instances, removing static 14-day cap to optimize recommendation quality and storage costs • Empowered all dev and QA teams to debug song recommendation issues by building internal React/JS website and Java/Spring service to search and filter user recommendation history across all production and testing environments ### Software Engineer Intern @ Pandora Jan 2021 – Jan 2021 • Improved speed and efficiency of radio station shuffle feature by utilizing Redis to cache computed station sequence. Verified consistent user experience with AB test. • Allowed for use of advanced song tag data within recommendation pipelines on demand by creating service in Java to efficiently interface with REST API and log analytics for service. • Created a more natural listening experience for users by designing new method to intelligently control favorited song boost within radio stations and across listening sessions. ### Software Engineer Intern @ Truepill Jan 2018 – Jan 2018 | San Mateo, CA • Created a custom algorithm which automated financial data collection system, making it quicker and easier to find relevant metrics on demand. • Created a script which aggregated relevant shipping data to better understand how shipping costs varied week to week along with data insights on a per medication level. • The software I built is now being used weekly by the operations team to automate the reporting workflow that previously took members of team several hours to complete manually each week. The reports generated by my code are then viewed by several members of the company on a weekly basis, including senior leadership ## Education ### Master of Science - MCS in Computer Science University of Illinois Urbana-Champaign Jan 2023 – Jan 2023 ### Bachelor of Science - BS in Computer Science University of Illinois Urbana-Champaign Jan 2019 – Jan 2022 ### PSEO in Computer Science University of Minnesota Jan 2018 – Jan 2019 ### Edina High School Jan 2016 – Jan 2019 ## Contact & Social - LinkedIn: https://linkedin.com/in/sujit-varadhan --- Source: https://flows.cv/sujitvaradhan JSON Resume: https://flows.cv/sujitvaradhan/resume.json Last updated: 2026-04-01