Back to browse
GitHub Repository

An AI-powered research assistant that helps researchers, students, and curious minds find, understand, and organize scientific literature faster.

7 starsTypeScript

Building an open-source Google Scholar / Consensus alternative

by nexwastaken·Feb 12, 2026·2 points·0 comments

AI Analysis

●●SolidShip ItNiche Gem

Full-stack learning project that actually solves a real problem for researchers.

Strengths
  • Complete feature set: semantic search, AI summaries, rich-text notes, sharing, folder organization—not just a wrapper.
  • Ambitious architecture: monorepo with Next.js frontend, backend services, analytics, payments, mobile/extension groundwork.
  • Open-source with active development and willingness to add BYOK and multi-model support based on feedback.
Weaknesses
  • Only 5 stars and 1 fork suggest limited traction; unclear if Semantic Scholar API integration is reliable or sustainable.
  • Competing against established players (Google Scholar, Consensus, SciSpace) with larger teams and funding—differentiation unclear.
Category
Target Audience

Researchers, students, academics seeking alternatives to Google Scholar and Consensus

Similar To

Consensus · Google Scholar · SciSpace

Post Description

I wanted to build a full-stack app from start to finish to learn the modern dev cycle so I built something that'd be helpful for me in research.

Started with a single Next.js app, switched to Monorepo due to growing complexity. Has analytics, email, payment, ai, backend, frontend, help & docs - everything most modern apps need. Even some barebones for a mobile app and a browser extension (using wxt).

The main features are a search engine powered by Semantic Scholar, AI summaries, folders - with rich text notes (using WYSIWYG) & sharing.

Thinking of adding different AI models and a BYOK feature.

Here's a quick demo video too - https://drive.google.com/file/d/1yv75PEghFN0-53M_Gb4EMefX-sQ...

Similar Projects

Productivity●●Solid

Joria – a native Mac notes app for instant capture and semantic recall

Instant-launch capture bound to a keystroke and local-first storage are the app's best ideas — you can jot something down without breaking flow and then query your notes by meaning. The 'chat with your notes' and use-your-own-AI angle is sensible for privacy-minded users, but the pitch leaves open how sync, model routing, or integrations actually work, and the space already has stiff competition.

SlickCozySolve My Problem
nanbing
104mo ago
Productivity●●Solid

A note-taking app where notes connect themselves

Unnamed drafts and automatic resurfacing are the two gutsy choices here — start writing and the app tries to pull up related notes for you. It's a neat UX shortcut if the semantic matching is strong, but the landing page hides how recall is implemented (model, privacy, offline), and in a crowded space with Mem/Reflect clones that detail matters.

SlickSolve My Problem
yamatokaneko
104mo ago