Inspiration
Writing academic papers often requires continuous guidance from supervisors, yet real-time feedback is rarely available while writing. This project was inspired by the need to bridge that gap by embedding an intelligent academic advisor directly into the writing environment, allowing researchers and students to receive instant, paragraph-level feedback as they write.
What it does
The project is an AI-powered Google Docs add-on that acts as a virtual research supervisor. When a user clicks on a paragraph, the add-on analyzes the text and provides plagiarism checks, advisor-style comments, suggested improvements, academic justifications, and relevant citation recommendations from online journals. This enables writers to refine their work immediately without leaving the document.
How we built it
We built the add-on using Google Apps Script for integration with Google Docs and a backend AI service powered by large language models and NLP techniques. The system sends selected paragraphs to the AI engine for analysis, retrieves academic feedback and journal references, and displays the results in a contextual popup or sidebar within Google Docs.
Challenges we ran into
One of the main challenges was ensuring that feedback remained academically accurate and context-aware for different sections of a research paper. Integrating real-time analysis without disrupting the writing flow was also challenging, as was balancing detailed feedback with concise, actionable suggestions.
Accomplishments that we're proud of
We successfully created a seamless paragraph-level interaction model that feels like real supervision inside Google Docs. The add-on delivers multi-dimensional academic feedback—plagiarism, reasoning, improvement suggestions, and citations—in a single interface, significantly improving the efficiency of academic writing.
What we learned
We learned the importance of contextual understanding in academic writing support and how critical user experience is when embedding AI into everyday tools. The project also deepened our understanding of integrating AI systems with productivity platforms while maintaining clarity, reliability, and academic integrity.
What's next for AdvisorAI: AI-Based Research Paragraph Writing Assistant
The next phase of AdvisorAI focuses on expanding its intelligence, usability, and academic impact. We plan to enhance context awareness so the system understands not only individual paragraphs, but also their roles within entire sections and chapters, enabling more coherent, section-level guidance.
Log in or sign up for Devpost to join the conversation.