Modern businesses run on spreadsheets. Can AI handle financial spreadsheet construction?
We’re excited to share MBABench 2.0: 101 difficult, long-horizon tasks (take up to 33h) across 6 finance areas and most financial model types.
The best AI agent passes only 16% of the tasks
Early in my MBA: AI wrote production code but couldn't build decent Excel models. I took this to Prof. Hong. He said SWE-Bench helped AI do great at coding; business had no benchmark. So we built MBABench: LLM-judge that grades financial models like a senior analyst. More soon!
Today we're releasing MBABench 1.0: the first, spreadsheet-focused benchmark evaluating AI agents on building entire workbooks to professional standards.
In the real-world, MBAs must produce complete, well-organized artifacts ranging from spreadsheets to presentations. These deliverables are then reviewed and revised by multiple stakeholders ranging from managing directors to client analysts, so beyond accuracy, they must meet various quality standards specific to the industry.
We evaluated 10+ frontier models on end-to-end spreadsheet tasks, and found agents still fall short of the standards MBA grads are held to. (1/n)
@WhatsApp I’m not able to change my number from my Indian number to a US number. I’ve raised multiple support requests but none of them have been resolved. Please help.
@visiblecare I have an issue with activating my number
Had a live chat with an agent, she tried to fix it. I tried what she told me but faced the same issue again. She asked me to escalate incase I faced it again. Please help.
2 incredible years at Zepto has ended. Working in a fast-paced company & witnessing its phenomenal growth firsthand has been an enriching experience. I learned more than I ever imagined, & I'm grateful for the opportunities I had there.
Next - MBA at Columbia Business School :)
Chatted with @airtribe_live about my shift from engineering to product management.
Shared some details about my projects at Ola too!
Check out the video: https://t.co/SlTxfOsI4C
#CareerJourney#ProductManagement
The day you became a clearer thinker, you:
-started by identifying the real goal
-decomposed vague concepts
-framed the right questions
-sought more data or experience
-listened to multiple perspectives
-assessed upsides & downsides
-examined your own biases
-acted like an owner