Ben Tomlin

bentomlin101@gmail.com (510) 362-5617
bentomlin.dev LinkedIn GitHub

Education

University of California, San Diego

B.S. in Math - Computer Science, 3.33 GPA, 6x Provost Honors

Sep 2021 – Mar 2025

Relevant coursework:

  • Data Structures and Algorithms in Java, C, C++, Assembly
  • Software Engineering and Web Development (Agile/Scrum, HTML/CSS/JS fundamentals)
  • Database Systems and Relational Theory (SQL, normalization, query optimization)

Experience

Founding Engineer at Tramona

React, Next.js, PostgreSQL, AWS

| Site| LinkedIn

San Diego, Oct 2023 – Dec 2024

  • Led a team of 5 full-stack developers using React, Next.js, and PostgreSQL to create a budget short-term rentals platform from the ground up, scaling to 30,000+ properties, 20,000+ users, and $300k+ in gross booking volume.
  • Introduced, documented, and customized a UI component library using shadcn/ui, TailwindCSS, and Radix Primitives, enabling a 45% reduction in frontend code size, faster development velocity, and more consistent designs.
  • Used TypeScript, PostgreSQL, NextAuth, and Zod to create a secure REST API and scale the backend to 250+ endpoints, and integrated React Query on the frontend to reduce API calls by 30% while improving maintainability.
  • Designed the PostgreSQL schema and evolved it using expand-and-contract migrations, scaling to 50+ tables and 480+ columns.
  • Leveraged Puppeteer, cheerio, residential proxies, and AWS Lambda functions to scrape 30,000+ properties across 20+ websites.

Projects

Volly AI (Full Stack)

React, Docker, PostgreSQL, AWS, OpenAI, OpenRouter

Mar 2025 – Present

  • Designed, built, and deployed an AI chat application with offline support using React, Docker, PostgreSQL, AWS Lambda, and AWS Fargate, allowing for next-frame interactions and instant search across all chats, prompts, and AI responses.
  • Engineered a RAG system that enables performant document search by incrementally computing and locally storing embeddings for prompts and responses, reducing backend load by 90% while maintaining search accuracy.
  • Leveraged OpenRouter to create a unified interface for 300+ AI models, web search, prompt caching, and image analysis.

MCSR Ranked (Frontend)

Svelte, TailwindCSS, D3.js

| GitHub| Site

Mar – Jun 2023, Dec 2024 – Present

  • Designed and wrote copy for the landing page, and built and maintained interactive statistics pages for a Minecraft speedrunning website, scaling to 200,000+ monthly page views and 10,000+ monthly users.
  • Migrated from a chart library to a custom D3.js solution to create a responsive, interactive, and performant graph to visualize up to 3 megabytes of data, boosting render performance by 95% and preventing freezes on lower-tier devices.
  • Leveraged IndexedDB to create a syncing and caching system, reducing API calls by 60% while enabling faster page loads, instant backwards/forwards navigation, and offline functionality.

donequick (Full Stack)

Svelte, Firebase, TailwindCSS, noSQL

| GitHub| Site

Aug – Sep 2023

  • Used Firebase to create a realtime lists app with optimistic updates, reordering, infinite nesting, and keyboard shortcuts.
  • Used Firebase Firestore and Firebase Auth to sync user data to the cloud and enable secure list sharing & collaboration.
  • Leveraged Firebase Security Rules to support secure list sharing and client-side database queries.

UCSD Courses (Frontend)

Svelte, Python, Web scraping

| Scraping GitHub| Site GitHub| Site

Jan – Mar 2023

  • Utilized Python and BeautifulSoup to scrape 179 majors and 6,888 courses from 82 UC San Diego department pages.
  • Wrote a Python script to parse, clean, and interlink the courses, and to generate HTML pages for each department.
  • Designed and created a new course catalog website using Svelte and TailwindCSS that lets users search departments, share links to courses, and click any mention of a course to jump to its description, creating a more efficient and cohesive user experience.

Technical Skills

Languages: JavaScript, TypeScript, HTML, CSS, Python, Go, Java, C, C++, SQL, Bash
AI Tools & Concepts: Hugging Face, Langchain, MCP, OpenAI API, OpenRouter, Fine tuning, Vector databases, Prompt caching, RAG
Frameworks & Libraries: React, Next.js, Svelte, Vue.js, Angular, TailwindCSS, shadcn/ui, D3.js, Chart.js, Node.js, Express, Flask, FastAPI, Redux, React Query, tRPC, Zod, Prisma ORM, Drizzle ORM
Other Tools: Git, GitHub, Docker, AWS, Firebase, Vercel, Netlify, SST, VSCode, Vite, npm/yarn/pnpm, Jest, Cypress
Save as PDF