By the analyzed data i create top 3 stock for next data trading
I use two factor to analyze
1.Technical analyses
2.Sentiment analyses
Technical analyses :
I use market technical data value to find future prediction use past data Sentiment analyses:
For the top 80 stock gather from Technical result i do sentiment analyze using finbert model with newsapi to find that stock relevant news and set a score on that stack depend on news type
By combining this two factor create top 3
Currently i am building this plan as SaaS product for RS299/month RS599/3month
I include AI fetaure using regular LLM AI chat which respond question by analyze AI pipeline result both technical and sentiment
SO what do you guys think about it?
How do you see this product as user?
I dont want to promote my product i want feedback formy product working principle and core architecture and i want to check real user really need a product like this now
we include backtest feature in product so user can know model accuracy
dont judge that accuracy is below 50% and other values are low currently its just a MVP still building AI pipeline for add more accuracy. currently i need to know the demand of this product in current market and can i get user base for this product
Hopefuly the first few hundreds of people paying 299 per month do not have significant capital to deploy in thhe markets and invalidate the patterns that it relies on.
I dont prefer blackbox strategies, I am distrustful of all 3rd party strategies, and I would lose conviction during DD. But I suppose there is a market for such if you can show past history (blind backtests with tortured data is always a danger though).
Start posting the findings here on an EOD bases for a few months and then maybe people will start getting interested in your app.
Well, see this guy, he is committed to posting this voluntarily everyday for the last 3 years or so for the benefit of everyone here. This is a model you can follow.
Two changes were made to prevent all users from buying the same stocks at the same time:
Higher liquidity filter — only stocks traded ₹25cr+ per day are eligible, so even mass buying won’t move the price.
Cohort splitting — the top 15 picks are divided into 5 groups of 3; each user is assigned a group based on their ID, so different users see different picks.
Net effect — instead of 100% of users buying stock A, only ~20% do, spreading the crowd across 5 stocks instead of 1.
This is exactly the kind of feedback I was looking for — thank you! Building trust through consistent free value before asking anyone to pay for it makes complete sense.
