Still figuring it out.

productHunt

github

Sahil Khan  • Peerlist

Research can be frustrating. Google Scholar has great functionality, but the UI sucks. Consensus is great, but their pricing can be a bit expensive. So I decided to built my own open-source alternative - Qurious. It's a all-in-one platform discovering, summarizing and organizing your research papers. 100% open-source.Features:<ul><li>Search across 250 million papers, powered by Semantic Scholar. Then summarize them quickly with Generative AI.</li><li>View detailed insights about every paper - citations, references, recommendations, and even a beta graph view.</li><li>Organize your papers into folders. Add your thoughts and notes (with rich text!). Then share it with the world!</li><li>No recurring subscription. Buy credits that never expire. </li></ul>⭐Star us on GitHub if you find the project interesting! ~ <a target="_blank" rel="noopener noreferrer" class="break-all link-break text-gray-500 hover:underline" href="https://github.com/QuriousAI/qurious">https://github.com/QuriousAI/qurious</a>

QuriousAI

- A GraphQL based backend API that tracked the ranked 1v1 and 2v2 queue for Brawlhalla. - The server was highly configurable with the help of config keys and offered real time updates through websocket subscriptions. - It also allowed clients to track (a.k.a. "snipe") whenever a specific player or team would start queuing (first of it's kind technology in the community). - The server was mainly used to power hornstat (the second largest bot in the community) and other internal bots. - Tech stack: GraphQL, TypeGraphQL, Websockets

Brawlhalla Ranked Queue Server

- Created a data pipeline using cron jobs, node JS, mongo DB to periodically push data from a third party API into a NoSQL DB. - Performed data analysis such as rating growth rate, rating prediction, member utility prediction and displayed results though graphs/bar diagrams (Jupyter Notebooks, Google Sheets). - Tech stack: NodeJS, Python, Google Appscript, Jupyter Notebooks

hornstat | Data Pipeline

- Discord bot for showing real time player statistics and game info. - The bot had the second largest market share in the community and was used primarily by all gamers to check their stats and other realtime updates. At it's peak the bot was handling 1000+ commands/hour. - Tech stack: Python, Node.JS, GraphQL, MySQL, Redis, AWS.

hornstat | Discord Bot

- It is a desktop application used for managing the team wins, losses and game history in seasonal leagues. - The application was responsible was storing each season with it's own teams, match history and leaderboard scores. - It was used mainly by the team staff internally to update the scores and generate leaderboards updates after a match. - Tech stack: Electron.js, Vue.js, Quasar.js

Leaderboard Management Application

- A web app to funnel image results from the midjourney bot. - Developed a discord bot to proxy results from the midjourney bot to a GraphQL API for consumption over other platforms. - Wrote a frontend to interact with the GraphQL API to generate images based on an inputted prompt. - Tech stack: Discord.js, GraphQL, TS, Vue, Tailwind

Midjourney Proxy

- A CLI tool which allowed tournament organizers to automatically seed their attendees (single player or teams) in accordance with their in-game ratings. - Reduced of hours of seeding time to seconds and was used to seed 10+ tournaments spanning over 500+ combined attendees.