2 ms·

I’m building a watchlist for tracking pretty much any kind of datapoint - stocks, weather, air quality, macro indicators, and more: https://curio-six-wheat.ver

by vwake 20d ago

I’m building a watchlist for tracking pretty much any kind of datapoint - stocks, weather, air quality, macro indicators, and more: https://curio-six-wheat.vercel.app/ https://curio-six-wheat.vercel.app/ It started as a curiosity while I was exploring some economic data. I wondered if there was a simple way to keep an eye on a particular datapoint over time, which led me to the broader question: could I have one watchlist for all the different numbers I’m interested in? It started out as a weekend project. I’ve been finding it surprisingly useful to use myself these days, and I’m continuing to explore where it could go. Would love some feedback on it from the folks here!

amelius 20d ago
Do you detect correlations? Do you do forecasting?
- vwake 20d ago
  Right now its limited to identifying the right data point and tracking it. I've recently added a followup feature which allows comparing with peers or with past values. I want to explore both correlations and forecasting! Can you share more about where you feel they'll be useful? Thanks!
gamebak 20d ago
Please make it work without a login, at least offer some free to use searaches so that we can see how it works.
- vwake 20d ago
  Makes sense. I added login in order to reduce LLM calls and search cost as this is just a side project. I'll add it after a few searches. Thanks for the feedback!
  - vwake 19d ago
    Now you can try it out without providing your email ID :)
laylower 19d ago
Are you considering making all charts extractable? So if I wanted to use a chart I could? Also, it seems like a RAG on top of an LLM. Is that right?
- vwake 19d ago
  On making charts extractable - that's a good point. I haven't thought about it so far. Curious about what you'd want here: the chart as an image, an embed, or the underlying data as a CSV? On RAG - not quite. All datapoints are fetched from trusted sources via API or web search. An LLM interprets the question and then routes it to a source. I am not sure if RAG could fetch live datapoints.

by vwake 20d ago

amelius 20d ago

Do you detect correlations? Do you do forecasting?

vwake 20d ago
Right now its limited to identifying the right data point and tracking it. I've recently added a followup feature which allows comparing with peers or with past values. I want to explore both correlations and forecasting! Can you share more about where you feel they'll be useful? Thanks!

gamebak 20d ago

Please make it work without a login, at least offer some free to use searaches so that we can see how it works.

vwake 20d ago
Makes sense. I added login in order to reduce LLM calls and search cost as this is just a side project. I'll add it after a few searches. Thanks for the feedback!
- vwake 19d ago
  Now you can try it out without providing your email ID :)

laylower 19d ago

Are you considering making all charts extractable? So if I wanted to use a chart I could? Also, it seems like a RAG on top of an LLM. Is that right?

vwake 19d ago
On making charts extractable - that's a good point. I haven't thought about it so far. Curious about what you'd want here: the chart as an image, an embed, or the underlying data as a CSV? On RAG - not quite. All datapoints are fetched from trusted sources via API or web search. An LLM interprets the question and then routes it to a source. I am not sure if RAG could fetch live datapoints.