Phind
AI
ai search engine
SHOW HN2023
1,401points
414comments
#3of 3,109 in 2023
Top 1%of 2023 by points
How HN took it
DebatedReal support and real criticism
58 positive, 66 neutral or asking, 16 mixed, 82 critical, out of 222 comments about the project. Thread score -0.09.
Execution: 25 for, 58 againstIdea: 14 for, 5 againstDesign: 5 for, 5 againstPricing: 4 for, 3 againstPerformance: 0 for, 6 against
The thread, comment by comment
Each row is a top-level comment and its replies, in HN's order. Select a cell to open that comment on HN.
positive neutral or mixed critical about another commenter or something else the maker
What HN said
What people liked
Good answers on recent libraries and niche technical questions
From 3 comments
Passes my smell test, which is to ask "how do I migrate my swift composable architecture project to structured concurrency". This uses 2 things that GPT-4 doesn't know about yet:…
afro88 on HNI tested it out and got some pretty good results - marginally better than GPT4, which is a high bar! It strikes me that we’ve been clamoring that a better Google needed to exist,…
johnfn on HNI have been using phind on and off for a few months. I found it amazing for discovery of software libs for a project I was working on. I could not find the libs when searching…
supermatt on HN
Useful for daily developer searches and discovery of libraries
From 2 comments
I've replaced 90% of my Google searches with Phind in the last few weeks. My use cases are learning a new API, debugging, generating test cases. It's amazing. Real time saver.…
dalmo3 on HNVery impressed. Seen a lot of AI stuff coming out but this is: * Fast * Works even with HN hug of death * Useful for my daily flow (I mix chat gpt with google searches) I tried…
quickthrower2 on HN
Objections
Hallucinated or outdated answers on some queries
From 4 comments
I asked it this question[1], ``` I traverse a maze using a basic A* implementation (using the Manhattan distance metric). However, after the traversal, I would like to find out…
dvt on HNI always test with a query about Django "not equal" filtering and it always hallucinates, same thing here (the "ne" lookup doesn't exist and has never existed): To do a “not…
weird-eye-issue on HNIt seems to work nicely on simple queries, however there are some rough corners which I don't think have a simple solution. For example the query "how to set the timezone in…
Etheryte on HNNice - I've used the product before but noticed it sometimes gives hallucinated answers if I ask something for which there's no good google result. Is this something you plan on…
jasonwcfan on HN
Unclear monetization and sustainability
From 3 comments
So, who pays for this? This seems to be extremely expensive and somebody is footing the bill — this immediately raises questions as to how sustainable this is. I'm worried that I…
jwr on HNWhat’s your business model for this? Free GPT-4 seems too good to be true… whats the catch?
jonplackett on HNThis worked well for the one sample query I tried. Running unlimited GPT-4 API calls (plus search API calls maybe?) for people sounds expensive. What is your monetization…
illnewsthat on HN
Expert mode gains are unclear compared with default mode
From 1 comment
I've replaced 90% of my Google searches with Phind in the last few weeks. My use cases are learning a new API, debugging, generating test cases. It's amazing. Real time saver.…
dalmo3 on HN
Requests
Share link for answers
From 1 comment
I wish we could have a link to share the answer, really useful when it involves code snippets and explanation. - for chatgpt/expert mode
lastdong on HN
Support for PDFs and papers as sources
From 1 comment
I just tried this on questions I has about archery and bow design. It was immediately useful in highlighting and summarizing sources into something coherent while citing sources…
hosh on HN
Revenue sharing with content producers
From 1 comment
I've seen a lot of people saying this is the future of search... But this is so destructive for content producers, why would they continue to publish content that has no chance…
l2silver on HN
Questions
How the web content is retrieved and embedded
From 2 comments
Passes my smell test, which is to ask "how do I migrate my swift composable architecture project to structured concurrency". This uses 2 things that GPT-4 doesn't know about yet:…
afro88 on HNIt's a good step up from using pure GPT4. What do you think how they built it? How do you access content from other pages so quickly? My guess is that they have crawled a lot of…
ck_one on HN
Whether GPT-4 was used before this launch
From 1 comment
I've replaced 90% of my Google searches with Phind in the last few weeks. My use cases are learning a new API, debugging, generating test cases. It's amazing. Real time saver.…
dalmo3 on HN
Alternatives named
ChatGPT
From 2 comments
I just tried this on questions I has about archery and bow design. It was immediately useful in highlighting and summarizing sources into something coherent while citing sources…
hosh on HNOn Expert mode, I decided to ask it a simple question but in a niche language, to see how well it can scour the internet. ``` How do I emit JS object literals in a ClojureScript…
koito17 on HN
Google search
From 1 comment
I've replaced 90% of my Google searches with Phind in the last few weeks. My use cases are learning a new API, debugging, generating test cases. It's amazing. Real time saver.…
dalmo3 on HN
Kagi
From 1 comment
Kagi, for me
esafak on HN