Hacker Newsnew | past | comments | ask | show | jobs | submit | rallies's commentslogin

https://rallies.ai

Working on building an investment assistant backed by real time data. ChatGPT and Perplexity finance are amazing, but all of them are based on web search data only, which is a big limitation in finance since realtime data is important.

We have an agent that has access to almost every data point you can think of in the stock market (as much as we can get), which gets leveraged before answering.

And we also figured out ways to build amazing charts in between answer snippets, which looks very cool. Investors are usually very visual.


I work at the intersection of AI and investing, and I'm really amazed at the ability of this model to build spreadsheets.

I gave it a few tools to access sec filings (and a small local vector database), and it's generating full fledged spreadsheets with valid, real time data. Analysts in wallstreet are going to get really empowered, but for the first time, I'm really glad that retail investors are also getting these models.

Just put out the tool: https://github.com/ralliesai/tenk


Can't wait for being fired because some VP or other manager asked some model to prepare list of people with lowest productivity to pay ratio.

Model hallucinated half of the data?! Sorry we can't go back on this decision, that would make us look bad!

Or when some silly model will push everyone to invest in some radicoulous company and everybody will do it. Poisoning data attack to inject some I am Future Inc ™ company with high investment rate. After few months pocket money and vanish.

We are certainly going to live in interesting times.


That's more of a management problem than an AI problem. You could get the same result by replacing "model" with "intern" or "dude from Fiverr".


With one important difference: nobody would be able to tell if you did the spreadsheet or AI spew it. And you do not pay for that one specific task to be done out of your pocket.


Here's a nice parsing of all the important financials from an SEC report. This used to be really hard a few years ago.

https://docs.google.com/spreadsheets/d/1DVh5p3MnNvL4KqzEH0ME...


Doesnt SEC provide XBRL data and the statements in excel?


Nice tool - I appreciate you sharing the work!


We're running some live experiments these days, for both stocks and options. https://rallies.ai/arena


With actual money? Or still fake money?


Fake money is better than nothing, but one hopes that at the very least they’re correctly managing prices with the bid-ask spread, although real money would tend to influence what the actual numbers would be (small dollar amounts likely getting worse pricing, large dollar amounts potentially impacting the movement of the market).


If they are not actually trading, they are almost certainly getting a lot wrong and I would not trust this one bit.


I think the big limitation of nof1 is that they're not using a lot of data that an actual investor would use when researching companies.

We're trying to fix some of those limitations and run a similar live competition at https://rallies.ai/arena


This is pretty cool.

We're also running a live experiment on both stocks and options. One difference with our experiment is a lot more tools being available to the models (anything you can think of, sec filings, fundamentals, live pricing, options data).

We think backtests are meaningless given LLMs have mostly memorized every single thing that happened so it's not a good test. So we're running a forward test. Not enough data for now but pretty interesting initial results

https://rallies.ai/arena


Is the code/prompts used open source? if not how can we say it's ligit


How is Qwen so much worse than the rest (for the period accounted)?


Developer here: We've only started this last week so not enough data right now, but it's a really interesting experiment


Working on building a chatgpt wrapper with real time stock market data. More than 70% investors are using Chatgpt for their investment analysis these days, but the data is quite dated since it's all based on web search. Trying to fix it.

https://rallies.ai


Correction: made a mistake in assuming equal contribution for all holdings. Just fixed it to account for the actual value. The correct returns are 90%.


Came here to say this. Although the UI is clean, it's in no way a great user experience using NotebookLM. It's just such a great product so I go back to it, but the user interface is not my favorite part.


So that's 10% of their latest series?


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: