Project Info

LeXcel

Devpost

Inspiration

Spreadsheets are powerful but often frustrating tools. They require users to memorize complex formulas and spend time on tedious formatting, leading to a slow and efficient process. We wanted to create a solution that enhances productivity, reduces errors, and makes spreadsheets smarter and more intuitive.

What it does

LeXcel is an AI-powered add-on that enhances traditional spreadsheets by: Simplifying complexity โ€“ Enables users to generate complex formulas using natural language prompts Eliminating tedious formatting โ€“ Structures and formats new data given user requests Iterating With Verification โ€“ Learn from previous mistakes to produce consistent and robust solutions

How we built it

We developed LeXcel as a seamless add-on in Google Sheets that is easy to download and run for anyone: no coding experience required. Our main tech stack is javascript, google script, sheetAPI, OpenAI, and Gemini.

Challenges we ran into

Ensuring AI suggestions are accurate and context-aware for different spreadsheet use cases Building a UI/UX that integrates seamlessly without disrupting existing workflows

Accomplishments we're proud of

Built a user-friendly AI-powered assistant that significantly reduces the time needed for spreadsheet tasks Successfully implemented real-time error detection and reasoning to fix its own errors

What we learned

There is a tradeoff between AI automatically inputting all formulas which may feel like magic, versus prompting the user for confirmation which could lead to more consistent results

What's next

Allow LeXcel to manipulate delete/insert rows, move data around, etc. Give LeXcel the ability to generate and execute multi-step computations Tackle more complex puzzles and Excel World Championship tasks Integrating with more spreadsheet platforms beyond Google Sheets such as Paradigm and Excel

Analysis

Compare with all teams

View

Metric

Figures cover GitHub contributors during the hackathon window. A co-authored commit counts in full for each author, so per-member totals add up to more than the whole-team figures.

Technology

Found in codeClaimed only
  • HTMLIn code
  • JavaScriptClaimed
  • OpenAIClaimed

1 of 3 appear in the indexed code. 2 claimed on Devpost could not be matched to code, which may simply mean the tool leaves no trace in the repository.

AI coding agents

No AI coding agent signals were found in this repository.

Detected from committed agent config files and commit authorship. Absence of a signal is not proof an agent was unused.

Codebase size

Source size

29 KB

Source files

2

Counts recognized source files only; vendored directories, binaries and lockfiles are excluded, so this is smaller than the repository on disk.

0 stars