hydershykh

This is a cool project, thanks for making it open source. Wonder if you've compared the results with ChatGPT by running the same queries.

That's a fair point. But models like GPT4 do not hallucinate much when it comes to summarizing. So I don't think these applications contribute to anything negative.

I think some of the financial applications around LLMs right now are better suited for things like summarization, aggregation, etc.

We at Tradytics recently built two tools on top of LLMs and they've been super popular with our usercase.

Earnings transcript summary: Users want a simple and easy to understand summary of what happened in an earnings call and report. LLMs are a nice fit for that - https://tradytics.com/earnings

News aggregation & summarization: Given how many articles get written everyday in financial markets, there is need for a better ingestion pipelines. Users want to understand what's going on but don't want to spend several hours reading through news - https://tradytics.com/news

Fair comment.

Retention during downtrends is pretty hard, so although we've gained a good amount of new users, old ones have left too.

3000 were paid users at that time, the total users were about 35k. We now have 54k total and about 3500 paid!

Started trading back in 2020 and realized how bad I was it, and how much I needed a strict system to eliminate emotions! Had a degree in CS so spent a few months reading up on quant related stuff and realized how negligent I and other retail traders were, and how much data driven insights could help (ran over a million backtests the last 2 years)!

Built a simple analytics tool to show dozens of correlations & analytics between a wide variety of financial data points, and bootstrapped it to $120k/month! We grew from 0 to 54k users in two years, all organic. Has completely changed my life.

tradytics.com

Developer of the extension here, happy to answer any questions.

Developer of the plugin here. Right now, this only works for twitter on firefox, but I plan on expanding it to other platforms like Facebook & Youtube and to other browsers like Chrome.

Here's a quick video on how it works: https://streamable.com/xo5x56

Happy to answer any questions.

Initially, just posted in a bunch of subreddits on reddit to get engagement. But the retention was quite low. I still used that engagement to get a couple hundred followers on twitter.

Afterwards, realized that twitter had the most solid trading community out of any other platform. So all I've been doing for the last one year is build a community on twitter now (and of ofcourse discord too since everyone's there these days). I try to post ideas from the system every day, and when people see them work and make some money off of them, they subscribe to the paid plan.

Discord has a massive portion of the overall trading community, so I created discord bots to pull data from the web product, and every data item had our logo on it. That helped quite a lot in expanding the brand name and reaching more people. We're in about 250 discords and reach about 100k members in them overall.

Finally, 12 hours of sitting in front of a laptop every day lol.

I have about 6 years of applied AI experience and I have been trying to work on something that can use that experience and be useful to others.

Ended up building an analytics based platform that uses data analysis techniques to generate trade ideas for retail investors [0]. Started it as a hobby July 2019, but it ended up growing quite big. The plan was to generate about $1k/mo but we are generating about $120k/mo these days with over 3000 paid users. I think I'm the most fortunate person in the world, never in my wildest dreams did I imagine this much.

I still work on it solo and have no plans on hiring a team, because solo is where I have most fun, and that's important.

[0] https://tradytics.com

haha I am the developer of it so I can't really say anything lol. Although we don't have OTC, a lot of other stuff is free on the website so give it a go with a free account. Always happy to answer questions.

Well, we might want to consider changing the criteria a little bit. No?

This is just crazy, someone doesn't like when the retail is winning. Looking at the options flow for this, more than 100 million dollars have been spent on contracts today. Halting it all of a sudden only makes people lose money.

Here are the options flow stats, crazy amount being spent on options. https://i.gyazo.com/dca500ef0f774a0019f0a42b7ac1e855.png

Source: https://www.tradytics.com

A one-stop shop trading toolkit to empower retail traders. I see so many traders lose shit ton of money just because they have no idea when it's NOT the time to buy. Been working on a platform to use AI to suggest users decent entry exit points. It's not totally automated but that's the point - use AI and humans together!

https://tradytics.com/

Writer of the blog post here. Happy to answer any questions you folks might have.

Thank you, working on a video right now.

I'll have a look at the link and see if we can add more.

Hence the comments on adding a lot more stuff.

I know it's not Quantopian right now but quantopian also started with a simple backtester which they then expanded quite well. I figured if there's enough interest, I'll just add on top of that idea and create something similar where you can backtest both with code and with just the UI.

Developer of the tool here. With the recent shut down of Quantopian, I thought why not start developing a similar system.

I have already been working on this for a few months and it just went live a few weeks ago - I wanted to make it a paid backtester but with the shutdown, I figured people would want different alternatives, so here it is. Completely free and offers many strategies for testing.

If there is enough demand, we can add a lot more stuff.

Happy to answer any questions.

I will try to modify it such that it's easy to add factors. Thanks for sharing this.

It does, all you need is to change your stock list to etfs.

It does not.

That should be doable pretty easily. The code is easy to read and modify.