AI-Powered Financial Modeling and Spreadsheet Automation

AI can reduce 4-5 hours of Excel work to 10 minutes

The agent handles everything from data extraction to formula correction, allowing users to supervise rather than manually build models

Why: Manual spreadsheet work is repetitive and error-prone; AI can process large datasets and complex formulas faster than humans

"it takes hours of work and makes it like 10 minutes"

Every data point must be traceable to its source

The system cites exact figures, PDF pages, and years for all extracted data, making every number auditable

Why: Financial models require accountability; users need to verify data sources before making decisions

"it can actually cite every single part of the information down to like the exact figure, the exact page of the PDF it came from"

Clarifying questions improve both AI performance and user prompting skills

Instead of requiring perfect prompts, the system asks clarifying questions which helps users learn to prompt better

Why: Most Excel users aren't good at prompting AI; interactive clarification creates a learning moment

"users are really not that great at prompting... when you show clarifying questions, it can make them better"

AI should enhance existing workflows rather than replace them entirely

Users can import/export Excel files seamlessly, maintaining compatibility while gaining AI capabilities

Why: Enterprises have existing tech stacks and processes; compatibility reduces adoption friction

"you can import and export Excel directly. So no one would ever even know it's in shortcut"

AI should find and fix its own mistakes in real-time

The agent detects circular references, formula errors, and calculation mistakes, then corrects them automatically

Why: Spreadsheet errors compound quickly; real-time correction prevents downstream issues

"it's finding its own mistakes and just chugging along"

Vague prompts encourage creativity and better clarifying questions

Less specific prompts force the AI to think creatively and ask better clarifying questions

Why: Overly specific prompts limit the AI's ability to optimize solutions; vague prompts surface better approaches

"I've always liked relatively vague prompts because it forces the frontier models to really get creative"

The best AI products feel obvious in hindsight

Successful AI tools solve problems so clearly that users wonder why they didn't exist before

Why: Obvious solutions reduce education costs and create viral adoption

"the best predictor of what that is, is if you can release something, do people say, I can't believe this didn't exist already?"

Humans will evolve from creators to supervisors of AI work

Finance professionals will supervise AI agents rather than build models from scratch, becoming 10x faster

Why: Supervising work is easier and faster than creating it; this shifts human role to quality control

"they'll progress to this role of supervisor, where they're 10 times faster"