Inspiration
Jeffrey and I are both undergraduates at UC Berkeley, and know how hard or overwhelming it can be to look find a summer internship. Like most of our peers, we also looked for research opportunities at our university, only to find postings decentralized, unorganized, and under-utilized. There was no one database to go and see all of the research labs at our school, or an easy way to find opportunities that interested us. There was also not an easy way to apply for these positions, even though most of them asked for the same information on each application. So, after hearing the CEO of Simplify speak at LAHacks, we got to thinking of a similar idea to help students unlock access to the experiences gained when working in research.
What it does
Papertrail is a 'one-stop shop' for research opportunities geared towards undergraduate students. We use AI to aid students in finding labs that they may never have thought to look for, and then assist them in every step of the application process. Our website is very intuitive and contains multiple features to help students land a position in research.
How we built it
PaperTrail was built as a full-stack web application using Flask for the backend and Jinja2 templating for dynamic HTML rendering. We started by designing core features like user authentication (registration, login, logout) with session management to personalize the experience. For the frontend, we emphasized a smooth and minimal UI, using HTML, CSS, and JavaScript, enhanced by frameworks like Flexbox for responsive layouts. The find_positions page uses a simulated database of research postings, displayed in a side-by-side interactive job board similar to LinkedIn or Simplify, with live detail updates when selecting a posting. The application_tracker page was designed spreadsheet-style, allowing users to add and update their application statuses with clean dropdowns and color-coded tags. To source lab and job data, we originally created a Python web scraper using BeautifulSoup to parse real research opportunities from the web. Throughout development, we focused on making the experience seamless, intuitive, and helpful for students navigating research opportunities at UC Berkeley.
Challenges we ran into
The biggest challenge was finding a dataset for the research labs or position listings that we could use to scrape data from. There is no single resource where all of these positions or labs are posted, so our AI agents had to search the web to find them. In some case, the data was outdated and the websites didn't exist anymore. For our success, we would want to work directly with these labs to get their postings from them and pipeline it into our product to create a employer-to-applicant portal where lab groups get to make posts about their opportunities and users can look through and apply to them.
Accomplishments that we're proud of
The web application is fully functional and includes all of the features we originally thought of when brainstorming this project. For less than 48 hours of work, we managed to create a full pilot site that is ready to demo, and could start getting users once hosted. The learning experience throughout this process was a huge demand, but Jeff and I split up the work well and had a vision of what we wanted Papertrail to be.
What we learned
Building PaperTrail taught us a lot about what it takes to bring an idea to life as a full web app. We learned how important it is to balance functionality with a simple, clean user experience. Throughout the process, we saw how much planning, iteration, and small details go into creating a platform that feels smooth and useful. We realized that building a great tool isn't just about code — it’s about understanding the people who will use it and designing something that actually makes their lives easier. Most importantly, we practiced thinking from a user’s perspective — creating features that are not just functional, but intuitive and helpful for real-world student needs. Funny enough, while creating PaperTrail, we actually learned about Berkeley labs in our department that we have never heard of before. We thought we had pretty much seen all the labs that Berkeley's EECS department sponsors, but to our surprise there were several that neither of us have ever seen.
What's next for PaperTrail
Jeff and I have been talking about taking this further. We were really inspired by Michael Yan's speech, and we're both interested in taking the startup route with this project. We have a fully functional pilot and have done some market research into the target demographics and how useful this application would be. When we get back to Berkeley, we're planning on talking to directors of research labs to seek interest in supporting our mission, and taking our product to entrepreneurship programs that can help us find investors and gain a support network for creating a company. We're really excited to see where this can go!
Log in or sign up for Devpost to join the conversation.