5 ms·

Show HN: We've built ChatGPT for your pdf files

13 points by alokjnv10 4y ago

My friends and I were really excited to try out ChatGPT when it was released. We were amazed by its power and capabilities. We were thinking how great it would be if people could use ChatGPT to ask questions from their own data. So me and my friends had decided to build such a tool. We are delighted to announce that we are releasing a public beta of our tool> After you upload all your pdf files into it. It can search for relevant documents without perfect keyword match, summarize takeaways from the document specific to your question, and extract key information from the document. It can help with brainstorming and summarization. We would love if you try it out and let us know what you think about it. We look forward to hearing your feedback on comments section below :)

elektor 4y ago
I just tried this out on a research abstract and asked it two softball questions (How many patients in trial? What was main study outcome?) and it got both of them right :)
- alokjnv10 4y ago
  good to hear that its working. Thanks for trying it out.
  - alokjnv10 4y ago
    we would really appreciate if you could share it on your twitter. @CorporaAI
- navdeep_dhakar 4y ago
  This is so great to hear, we soon will be working on improving our tool on task specific jobs like research paper analysis, reports, companies internal data ,chats etc.
- kevviiinn 4y ago
  If only researchers asked softball questions. I can't say it's an accurate use case
conaire 4y ago
Is able to compare text from multiple documents :)
- alokjnv10 4y ago
  yeah it works for that as well
navdeep_dhakar 4y ago
The possibility we saw with this tool was in so many usecases, but we started with reports,document, research paper etc. But we will be expanding it into lot more, like it will be aware of data context, more intelligent, multisource of data, different data type etc. Possibilities are enormous.
d4rkp4ttern 4y ago
I wish the landing page doesn’t talk about “employees looking for documents”. Why is it specific to “employees”?
- alokjnv10 4y ago
  we thought employees can be our target customer
  - replwoacause 4y ago
    No
d4rkp4ttern 4y ago
Neat. Is it really using ChatGPT? Asking because ChatGPT does not have an official API, but GPT3 does.
iosono88 4y ago
[dead]

13 points by alokjnv10 4y ago

elektor 4y ago

I just tried this out on a research abstract and asked it two softball questions (How many patients in trial? What was main study outcome?) and it got both of them right :)

alokjnv10 4y ago
good to hear that its working. Thanks for trying it out.
- alokjnv10 4y ago
  we would really appreciate if you could share it on your twitter. @CorporaAI
navdeep_dhakar 4y ago
This is so great to hear, we soon will be working on improving our tool on task specific jobs like research paper analysis, reports, companies internal data ,chats etc.
kevviiinn 4y ago
If only researchers asked softball questions. I can't say it's an accurate use case

conaire 4y ago

Is able to compare text from multiple documents :)

alokjnv10 4y ago
yeah it works for that as well

navdeep_dhakar 4y ago

The possibility we saw with this tool was in so many usecases, but we started with reports,document, research paper etc. But we will be expanding it into lot more, like it will be aware of data context, more intelligent, multisource of data, different data type etc. Possibilities are enormous.

d4rkp4ttern 4y ago

I wish the landing page doesn’t talk about “employees looking for documents”. Why is it specific to “employees”?

alokjnv10 4y ago
we thought employees can be our target customer
- replwoacause 4y ago
  No

d4rkp4ttern 4y ago

Neat. Is it really using ChatGPT? Asking because ChatGPT does not have an official API, but GPT3 does.

iosono88 4y ago

[dead]