Hacker News new | past | comments | ask | show | jobs | submit login
Show HN: Phind V2 – A GPT-4 agent that’s connected to the internet and your code (phind.com)
221 points by rushingcreek on Aug 7, 2023 | hide | past | favorite | 134 comments
Hi HN - Today we’re launching V2 of Phind.com, an assistant for programmers that combines GPT-4 with the ability to search the web and your codebase to solve nearly any technical problem – no matter how difficult.

We’re incredibly grateful for the feedback we received when we first launched GPT-4 answers back in April (https://news.ycombinator.com/item?id=35543668). As Phind has gotten better at complex programming tasks, the questions it gets asked have gotten more complex as well. In the past, we would always perform a web search for every input. This limitation constrained Phind’s answers to what was present in the search results, preventing us from making Phind a more powerful debugger and making it challenging to integrate Phind with your codebase.

We’ve addressed all these shortcomings in Phind V2. This release has three major updates: (1) Phind is now a pair programming agent that knows when to browse the web, ask clarifying questions, and call itself recursively; (2) the answering engine defaults to GPT-4, and you can use it without a login; (3) we integrate with your codebase via our new VS Code extension.

We realized that search is only one of many tools that Phind should be able to use. As such, Phind has been re-engineered to be an agent that can dynamically choose whatever tool best helps the user – it’s now smart enough to decide when to search and when to enter a specialized debug mode. Instead of making assumptions about your code and proceeding blindly, Phind can ask you questions and clarify its assumptions. When a problem requires multiple searches or logical steps to solve, Phind can call itself recursively and perform multi-step reasoning without user input.

We’ve heard from you that switching between your IDE and Phind in the browser has been a major pain point. No longer – we’re launching a VS Code extension that brings Phind into the IDE and finally connects Phind with the context of your codebase. Phind in VS Code automatically determines which parts of your code are relevant to your search and can help you squash bugs in a single click.

To maximize Phind’s alignment with your preferred answer style, we’ve also added a feature called Answer Profile where you can tell the AI about yourself. Phind will apply this answering style across the board, automatically.

Here are some examples of the new Phind answering questions it could not before:

Clarifying assumptions to help a user with debugging: https://www.phind.com/agent?cache=cljmjzjgn0000jo085otq111f

Designing a highly specific and custom database schema: https://www.phind.com/agent?cache=clkwpprz600g4jt08dl21e7r6

Splitting a Wordpress theme across multiple files: https://www.phind.com/agent?cache=clknqywuq001pji083sdacf9p

Phind’s asking clarification questions in debug mode: https://www.phind.com/agent?cache=cljmjzjgn0000jo085otq111f

Phind extension answering questions about a local codebase: https://www.phind.com/agent?cache=ra4kh2v3epgv5iw7z6dlzuo4

Answering questions about a local codebase using the web: https://www.phind.com/agent?cache=ztiaju6xwtpi39l2kjdnwh20

We are incredibly grateful for the feedback the HN community has given us and are excited to hear your thoughts about this release!

Cheers, Michael and Justin




I love phind. It has helped me a lot but the hard coded (or heavy weighted) ‘ask user question’ after every single query can be too much. Especially when it asks to see parts of the code that aren’t relevant to the task, and it does this is so much that I started to wonder if you were trying to scoop up more code blocks just to increase your data set.


The intention there is to suggest further directions to take your thread.


When the intention works then it helps. But as it stands it seems like an extra layer of fluff. If it’s primed to ask a question after every input, even when the instructions are clear as day then it automatically has to find a question to ask about, and will often ask irrelevant or questions that take you down the wrong path.

The type of question being asked, and also deciding IF a question even needs to be asked, needs some work. That’s all.


Makes sense. Thanks for the feedback.


Wow, I'm impressed.

I've been toying around with some prototypes to use AI and LLMs for hardware / chip design but at the HLS level using C++ rather than the HDL level using Verilog. This domain is just over the edge of what LLMs know about to spit out working, reasonable, and expert source code. This is the first public LLM/AI tool that has actually provided HLS code without needing extensive hand-holding.

Even more so, it is the best example I've seen that links high-level algorithms (from papers) or design concepts (from textbooks or online sources) to actual implementation (from documentation) without any concept based on the web access feature. In my own experiment with prompt engineering for hardware designs, I've always had to provide LLMs with a lot of context and they still often fail in this domain.

I will be toying around with this and I hope to see more in the future!

Edit: I just noticed the founders are in here. I am a PhD student and I would love to give detailed beta feedback in exchange for free credits to use this tool in my day-to-day research!


Thank you. Would love to know what we can improve.


Im a huge fan of this tool - I really like using it. I've started preferring it over google, and only googling after Phind fails me. Great work y'all! Keep it up :)


Instead of Googling, workflow seems to be for me Phind -> Kagi -> ChatGPT and just then Google. Times change.


Amazing -- love to hear it!


same I only Google if I need something Google excels at like place or maps or image search or video search, but anything deeper I go to phind. I find I'm thinking of search more in the context of a conversation than a diy try and find engine.


This puts everything I ask it on the internet. Is the intend to build something like an AI StackOverflow this way?

That should not be enabled by default.

If anything, it should ask the user "Are you ok with putting your question and the answer on the open internet?". And it should only be enabled for logged in users. And users should have a way to delete their conversations with Phind again.

It looks like there are already a bunch of conversations indexed by Google:

https://www.google.com/search?q=site:phind.com


Whoa what. This a game breaker. I’ve been using this tool for some personal side project stuff and had no idea that this was happening. Whoa. Not good.

To edit: I was never expecting full privacy as I assumed everything being fed into your service was being padded to your data set to make phind a competently fine tuned agent, but open web is not good.

In the end, the amount of times I’ve sworn at your bot with some foul language after it forgets everything on the 3rd chain and I have to go back and redo everything, will probably harm your seo if those swears get published. Just saying.


The cached link is currently shareable but won’t get indexed anywhere unless you yourself share it somewhere like Twitter. But we’ve heard the community’s concern about security through obscurity and we will address this shortly.


Are the 'cached' pages eventually deleted? I've seen some shared `https://www.phind.com/search?cache=*` URLs which redirect to the Phind landing page.

Wiping these pages would not be good from the preservation standpoint, especially since the URL doesn't seem human-readable.


So how hard would it be to brute-force to some share links then? I have been using this for a lot of searching.Should probably have read the ToS.

But to be fair, it feels way worse to use now than when it came out, so I guess I won't lose much by switching away at this point.

I hate this thing assuming it knows what I want.

Previously it had predictable behaviour, and acted like a slow LLM powered search engine.

Now I do a search and it starts asking me shit. I already tabbed away to three other places to do similar searches and come back to some half baked shit that has nothing to do with what I searched for. Plus I have to wait for it to go and do inference again before I can get to a link?

Why forget all the things that search engines got right over the years, but keep all the shit they got wrong?

https://www.phind.com/agent?cache=cll1bg5np0005l008molmt7d2

  import random
  import string
  import requests

  # Static prefix
  prefix = "cll1"

  # An empty set to store the cache links we've already tried
  tried_links = set()

  # Website URL
  url = "http://target.com/agent?cache="

  # Dictionary to store cachelink and its response text
  cache_link_dict = {}

  while True:
      # Generate a random 21 character base 36 encoded string
      suffix = ''.join(random.choices(string.ascii_lowercase + string.digits, k=21))

      # Combine the prefix and suffix to create the cache link
      cache_link = prefix + suffix
    
      # Continue with a new loop iteration if we've already tried this link
      if cache_link in tried_links:
          continue

      # Add the cache link to our set of tried links 
      tried_links.add(cache_link)

      # Make a request to the website with the cache link
      response = requests.get(url + cache_link)

      # Store the cachelink and its response text in the dictionary
      cache_link_dict[cache_link] = response.text

      # Add a delay between requests to avoid overwhelming the server
      time.sleep(1)

So it took 17 minutes to use your website to create the code needed to start enumerating these links.

Now given, I am pretty sure a large number of these will return nothing, but I am also willing to bet there are people that put GDPR data into these searches, perhaps a lookup of a phone number, or an address, names?

It would be quite trivial adding in the code to run these requests through a pool of proxies so they don't trigger anything too suspicious on your WAF.

I don't think I'm particularly skilled either, so if I can do this, I'm sure someone already has.


Theres ~10^31 possible links.

Assuming theres 100 billion valid links and someones scanning 1mil links per second, they'd still be averaging 1 discovered link per million years.


You can still use the basic search mode if you prefer maximum control.


I actually loved the original version when it just came out.

It beat ChatGPT every time simply because it unlocked a portion of ChatGPT locked down by the knowledge cutoff. It was also quite speedy, and even the way it resists going into infinite loops was much better than ChatGPT at the time.

I assumed my data might be used to train AI in an obtuse obfuscated kinda way, but I never would have imagined I could just brute-force cache links.


Why do you believe you can brute-force cache links? There could be some vulnerability but that string looks far too long to brute-force.


You are probably right, I was just sleep deprived and angered when I wrote that reply.

I am now only sleep deprived. I still use phind, and just came back here to say sorry to whoever, since it really does help me with productivity.

Also, it will require more time to reverse engineer the cache links, and I don't want to spend more time.

First all the requests return 403 so think there will need to be a selenium component (user-agent trickery is not sufficient).


fwiw I do like how easy it is to share conversations now. if you change it, please make the friction to sharing minimal.


I think that we'll made it private by default but with an easy "share" button you can click that opens it up.


You're right -- we need to have a way to make conversations fully private for logged in users. We'll work on this and will introduce more options for data privacy. Please feel free to reach out to me at michael@phind.com to chat further.

That said, no Phind links will be indexed unless you actively share the link to your conversation on an indexable site (like Twitter), which implies that you wish the link to be public anyway. I assure you that this is not an intentional dark pattern.


Not only for logged in users.

Even more so for users who are not logged in.

I find it very surprising that I type something into your service and that gets hosted on the open internet for everybody to see.

And as a logged out user, I have no way to ever delete it.

For logged out users, you really should not store the input at all.


No Phind links will be indexed or become easy to view for anyone else unless you share it on an indexable site (like Twitter), which implies that you're willing to make it public anyway. That said, we will add controls for this.


Not sure I believe in 'security by not finding the link'-- links get leaked through a variety of means that don't imply consent (e.g. browser history, metadata requests, or reshares).


I think what you have now is reasonable, considering the links are secret.

I am curious however how the Explore tab is populated? Are those conversations created by employees for demonstration purposes, or were they sourced from user interactions?


I mean, you should never type something on any free site you don't expect to be made public. That's an assumption on your part that is dangerous.


Really? You've never typed anything embarrassing into Google Search, like medical queries?


This isn't the posters line of reasoning at all, please don't derail the conversation with posts like this.


I find it very surprising that I type something into your service and that gets hosted on the open internet for everybody to see.

Did you expect anything else from an Internet-based cloud service? In fact, you just did that on this site, but I doubt you thought it's surprising?


He's logged in here and expressly conversing in public. You are comparing apples to nuclear reactors.


Until this happens and until you can verify that you’re not indexing every conversation to the open web, this is enough for me to full-stop using your service. I was under no impression whatsoever that the threads were made public.


We're not indexing every conversation to the open web. We've only seen this happen for threads that are deliberately shared publicly by the user. Please feel free to reach out to me personally at michael@phind.com and I'd be happy to discuss this further.


It’s a lot easier to start with privacy as a focus. It’s a lot harder to plug up all the holes after a public launch.

Looking forward to a lot of public updates! Repost back on HN when you all have resolved and I’ll start using the tool again.


This is frankly disgusting behavior. There's zero indication on the homepage that your questions and interactions will be published and indexed on Google, and I think it's an obvious expectation that they wouldn't be in the absence of informed consent. It's common to include proprietary code or sensitive business details when engaging with a tool like this, and it's a vantablack level dark pattern to make that information publicly available as an "SEO growth hack" without informing the user.

EDIT: It sounds like this might not be what is happening. @rushingcreek's original reply seemed to confirm it, and was only edited to add more information after I left this comment.


Nothing will be indexed unless you actively share the link to your conversation on an indexable site (like Twitter), which implies that you wish the link to be public anyway. I assure you that this is not an intentional dark pattern.


Chrome and other browsers use autocomplete to get websites to crawl.



The problem is that you create such links by default. I should not be able to open a link to a specific conversation and view that conversation 1. Unless I'm logged in and I have specific permission to view that item or 2. I have actively created a 'sharable' link.

People can type anything into these boxes, and if it's PII then you're on the hook for GDPR (not to mention the California equivalents) and implicit sharing and security through obscurity are not viewed favorably under those laws.


This is the current standard amongst AI search/answer engines and this is frankly the first complaint that we've heard about this.

That said, I hear you loud and clear and we will introduce more options for data privacy. Please feel free to reach out to me at michael@phind.com to chat further.


Neither ChatGPT, Bard, nor Claude behave in this fashion. To which ones are you referring?


If you use the ChatGPT "Share" feature and send the link to someone, anyone who has the link can view it. How else did you think it worked? They have no access control ui.


That is an explicit share. That link does not exist until you click the share button. With Phind, the URL to the chat is the share link. It's like every google doc having 'People with the link can view' turned on by default. If you create a random string and add it to their URL scheme and happen on a chat, you can view it regardless of if it's been intentionally shared by the creator, you don't even have to be logged into Phind.


Both Perplexity and You.com do this as well.


This seems to be a misunderstanding.

Which other "AI search/answer engines" are you referring to?


Perplexity apparently does this, I just confirmed it. Which sucks because perplexity is a very solid product that also lets you opt out of data collection and has a share button, implying action needs to take place to get it to be public


Yandex seems to have crawled many more results, some conversations are quite hilarious: https://yandex.com/search/?text=site%253Aphind.com


All of these links are examples that we ourselves have posted on our homepage.


Hey Michael and Justin, I just wanted to let you know that I rebuilt a small web app (used by ~2K people/month) with help from Phind V1, and I found it extremely useful.

The old app used Vue 2, PHP, and a dead build system. The new one uses Vue 3, Vite, and TypeScript (with serverless functions hosted on Vercel for now). This was the first time I'd used Vite and TypeScript seriously, and although I was familiar with Vue 2 it was my first time using Vue 3 for a project.

Phind saved me a ton of time (I'd estimate 50%) and hassle on the tasks I used it for. As an "enthusiast" software engineer, I asked it a ton of dumb questions that I couldn't possibly have asked on Stack Overflow, in Discord channels, etc.

Just so this doesn't come off as completely fawning, it wasn't unusual for it to be confidently wrong, in which case I'd correct Phind and try again. It was rare that I couldn't eventually get to a correct answer by correcting Phind's assumptions, missing information, or mistakes.

Best of luck to you and the team!


Thanks for the feedback! One of our main focuses for Phind V2 is reducing cases where it is confidently wrong by asking the user clarifying questions when it's stuck.


Oh wow. I've been using Phind and loving it, but I was thinking that you should integrate it with the editor. And then that's exactly what you did. This looks incredible.

But, I gotta say I'm a bit upset about the privacy issues raised above. It sounds like that was an honest oversight, so I hope it gets sorted out soon. I have to wait until then to use the extension. I think I can live with searches being out in the open (even if I don't like it) but I can't risk exposing any local code, even if it's supposedly impossible to find.


Noted. Thanks for the feedback.


I haven't heard a lot of buzz about Phind. But last time I tried it, it seemed better than Bing Chat or Bard or ChatGPT with search. Looking forward to trying V2!


++, having played with all three, agreed


I'm curious, is there a standard benchmark any one knows of that compares "practical usefulness" of LLMs instead of tries to make them take some kind of useless IQ test?

e.g. how useful is this LLM for 1) code debugging, 2) (accurate) fact retrieval, 3) daily task planning


Kagi did an evaluation a while back: https://blog.kagi.com/kagi-ai-search


Thanks! I love kagi's ethos!


One thing - Consider leaving the URL in your post blank so it routes to this HN discussion instead of your homepage. I completely missed this post of yours announcing V2 and its features until I clicked 'discuss'.

Rooting for you guys. I find myself using phind over other searches (google/chatgpt) more and more.


Ah thanks for the feedback!


Having used Phind for a while, I got quite frustrated by the new mode where it asks clarifying questions.

I think part of the frustration is that when I'm asking more obscure questions, which naturally have less hope of resolution, it would try to be helpful by branching out into questions that are often irrelevant.

Previously I'd get an answer that didn't satisfy my question, but at least some sort of hint of how far it got, with the option to continue from there, maybe using some terminology I was missing. Now I get a question as response, where I can't gauge whether it's reasonable to continue the likely dead end.

I'm often going to modify my original query like I would with a traditional search engine. Especially as the tool tends to forget earlier input and get stuck in the conversation and one solution. This approach of restarting is hampered by the tool going to ask for the "exact error message" even though I'd just copied that into my query and tried again.

So, I still like Phind for its summarising abilities, and will try how it integrates with a local repo, the conversational style has severe limits and should be toned down from what I see.


Gotcha, and this is specifically for debugging? It would be helpful if you could reply with a link to your query (or email it to me at michael@phind.com).


Hm, can't replicate currently. I'll let you know if it occurs again :-)


Love the concept. I hope there will eventually be an Emacs version of the extension - I couldn't imagine going back to VS Code.


This is on the roadmap! We want to be able to integrate with any and every code editor.


+1 for Jetbrains products.


Please do, I understand VSCode is a huge market but I'm too in love with my vim and my jetbrains stuff...


From the tutorial:

> In rare cases, this type of phrasing might result in irrelevant web results and therefore a bad answer. If you find that the web results are unrelated to your topic, try asking the question the same way you would on Google.

This seems to imply Phind isn't doing query expansion? (i.e. using GPT to generate several web queries)

Is this due to cost? (I looked into doing this myself and it would scale the costs by the number of queries.)


Ah the tutorial is a bit outdated. We do in fact do query expansion now in the default Phind mode (when basic search isn't toggled).


Sorry it says “please sign in to view plans”. Because of this, I cannot become a customer as that is extremely annoying UI


We've done this because it's not our goal to upsell you -- we only suggest you get a plan once you've exhausted your daily GPT-4 searches. And you can keep using Phind with the fast model (GPT-3.5) for free.


The price should be public, period. I had to create a dummy account just to view your pricing, which left unpleasant taste.


Wait, it's 10 per day, not 10 period? I used it last night and noticed that I had nine searches left with the "best model" before having to log in, and assumed they had to last me until the end of time. (By the way, after recently stopping using Phind because it frankly started sucking, last night's response was superb, A++, would (and will) use again.)


I absolutely loved using phind when it first came out that I was sharing it with all my coworkers and friends like I worked on commission. But about two months ago? during the chain integration, it went really downhill where I actually stopped using it completely. Really hoping v2 will draw me back in.


Hope so too! What are your thoughts so far?


I use gpt4 to write code (boilerplate stuff) and lately its been completely abysmal and can't remember prompts 3 chats back (next reponse does the thing i told it to not do 3 prompts back). Since you guys use gpt4 underneath, im guessing it has the same limitations. Have you guys noticed the same?


Can you try some of those same prompts in Phind? I’m optimistic that the answers will be better due to Phind’s internet access.


Still suffers from the same underlying issues (reasoning based limitations and lack of recall) (at least the stop option works on Phind). Unfortunately the performance tuning OpenAI has done the past months (theres a reason the 25 limit chat per 3 hours went away and there is no free lunch), created some sort of "uncanny valley" for responses now (it looks like what i want but not quite).

It lost the "damn thats exactly what i want" it had.


Would you mind pasting the link(s) to those Phind conversations or emailing me at michael@phind.com? Would love to dig in deeper.


I’ve noticed the same actually, it started about a month or so ago.


Congrats! Looking forward to seeing how it develops!

Here was an amusing one that just happened to me:

- query

- answer with hallucination-- 'try X' and citations

- sorry, that doesnt work

- oh, there must have been a mistake in the source material! please try Y

- works, thanks

I looked at the original source material, and the thing it hallucinated wasn't there at all :'). Nice try passing the buck GPT old buddy!

Edit: I'm still chuckling about this: "And finally, keep an eye on the Git version you're using. Some newer features like --no-attributes (although it turned out to be non-existent in this case) won't be available in the older versions."

https://www.phind.com/agent?cache=cll284giq000hmm098r51glbp


Thanks for including the link, we'll take a look at how we can improve this example.


I've found that the people that build AI tools and the people that use AI tools seems to be quite disconnected. Phind seems to be a notable exception, bringing some value over google. I am not entirely sure why it works better than goole tho. Is that because the model behind is better ? or just you figured that removing the usual trash (adds / over engineered seo) ? But i'll take it.

Now I am just concerned about the privacy of what I share (can it be worse than google ?) and the optimality of the code ? (code usually works, but could it work faster ? what happen when new libraries get out ?)


My only reason for not using phind is next 13. I use it for express and node js. But in the new app router it hasn't read the documentation and can't provide me with answers. In contentlayer with next it gave me a solution full of hallucinations and I spent a couple of hours implementing the solution it gave but later on I found that the issue was brand new due to new update.

An AI that says i don't know would be my next feature ask and that says using bing or something else with this search query. That would give me the peace of mind of not crosschecking to google too. But love it.

Please add support for jet brains/webstorm. I don't really use VS code.


pro tip, ask it to tell you something about next13 for the actual link, like can you verify that is the latest from nextjs.com? it doesn't always seem to do a search of it doesn't need to, if you explicitly ask it to first it will.

for example if you ask it give you lyrics in the style of x song, it will fail miserably but if you ask it what are the lyrics for x song then ask for similar lyrics it has the context. the key is making sure it has the context.

if you have an exact page to docs you want help with that's even better. sure it's an extra step but it's worth the time it saves you from reading and looking for examples elsewhere etc...

also I think some of the quality drop is on openai, their product has degraded some since May.


That's odd -- Phind should have done a search for next 13 and have gotten relevant docs. Do you mind pasting the link to that query or emailing me at michael@phind.com?


I have same issue with ChatGPT. Since it hasn't indexed anything post Sept 2021, I keep getting this message,

> As of my last update in September 2021, Next.js 13.0 has not been released yet. However, I can provide you an example of implementing authentication using the Next.js app router approach based on the information available at that time. Keep in mind that the actual implementation may differ in later versions of Next.js.

Does Phind do something special to get around this?


Yes, Phind is connected to the internet and can do web searches to get up-to-date information. For example, here's me asking about what's new in nextjs 13: https://www.phind.com/agent?cache=cll15y99k002fjo08ijfbedeo


That’s pretty neat, but can it answer with code suggestions with nextjs 13?

It didn’t work for me - https://www.phind.com/agent?cache=cll1y7pwf005vmh08j2zfyead


Do you monitor your `founders@` email inbox or is the dreaded Discord server the only real way to give feedback? Are you planning to create a public issue tracker... which can be indexed by search engines?


We have a public issue tracker in the Discord. And yes, we do monitor founders@


Is there any chance you'll create Intellij / IDEA extensions?


Yep, on the roadmap


As a heavy user of Webstorm I can't wait to see it!


I can’t sign in from multiple devices. It logs out of others if I do. Also, what will the pricing model?

OTOH, really like the new approach it is more human and tries to correct itself automatically.


That's odd, you should be able to be signed in on multiple devices -- will take a look.

As for pricing, you can view our plans at https://phind.com/plans. Our goal is to prioritize providing long-term value for you guys over short-term revenue, hence why we only suggest plans once users exhaust their free daily GPT-4 quota.


I really hope that locking pricing information behind a login is a bug at your end, because I'm not even going to consider a service that doesn't have up-front, clear pricing information. The very flowery MBA language you've used to side-step the question makes it seem intentional, however.

Why would users invest time going through your onboarding flow if they don't know whether they can afford the product, or what sort of value they should be expecting it to provide?


Fixing this shortly.


Happy Phind customer here (pro or whatever it’s called haha), but unable to use the VS code features due to not using VS Code.

Have you considered a LSP .. does the LS Protocol even support some functionality that you deliver?

In general I’ve been pretty happy with Phind. I hop back and forth between it and ChatGPT. Currently canceled my ChatGPT subscription in favor of Phind. Looking forward to future developments or perhaps generic implementations of the editor plugin for use in my own editors (Helix, if you’re curious).

Thanks!


Have you implemented any logic to prioritize official in-house documentation such as the software `man` pages and the output of the `--help` flag? I've gotten the impression that you've mainly weighted the Stack Exchange / Overflow pages in the results (while negatively ranking the information-wise ever-awful Quora and Amazon) and otherwise the scraper accesses the Bing's hits (or other search engine API you're using) in order.


Yes -- you can use https://phind.com/filters to customize your preferred sources.


Not feasible to entirely trim the results myself. Many programs host the documentation on their own domain and an absolute ton, especially smaller projects, don't have any online presence. I was wondering if you could maintain a database on Phind's end.


Have you had any issue with Phind not pulling up the right documentation?


1. I don't see any notice of this launch on the site.

2. The responses on phind tend to be long-winded, decomposing the question even when I try to formulate it logically and concisely.


We've quietly launched these features recently. As for response length, we have a feature called Answer Profile where you can tell Phind exactly how you like your questions answered (you must be signed in for this).


I can see why you would want to move beyond just googling the answer and summarising the relevant stackoverflow page.

But sometimes that's all I want, and it's a reliable way to save me 10 minutes. Reliability and ease of use are more important to me than the ability to answer advanced questions.

So I guess what I'm saying is be careful what you optimise for, and best of luck.


That makes sense, hence why we still left the old “Basic Search” mode available via toggle from the search page.


This is amazing.. Are you folks hiring ? :D


We are — please reach out at hello@phind.com!


Seems ok... https://www.phind.com/agent?cache=cll1botdq0025ky08w5z588zf

I would use.

But not after seeing that everything is made public. Can you enhance user privacy before I use please?


It's a misconception that everything is simply public. It's basically impossible for anyone to get your cached links unless you post it somewhere yourself. Perplexity and You.com do the same thing.

However, we will make links private by default for logged-in users.


You might want to sanction this particular user as the end of their transcript gets real classy... and by that I mean gross and racist.


It seems anyone can add further prompts to a thread.


phind is awesome, a few times it has helped me where GPT-4 could not due to its knowledge cutoff. some minor feedback on the UI, it doesn't always scroll to the bottom if it is generating code, and I have to manually scroll it into view


I tried using it but after one prompt it has stopped processing my further prompts. I don't get it... Is there a free limit or something?


I like Phind, but I feel like the quality of results I've been getting has been steadily dropping over the past few months.


We implemented some cost-saving measures in April/May, but since then we reversed course and this current product should be the best it’s ever been. Have you tried it recently? Particularly in the new default mode (not basic search)?


Responses to queries I made today seem quite slow compared to responses to queries I made back in March. Was that part of the cost saving measures?


Phind now runs on GPT-4 by default, even without an account. So it's quite the opposite -- the underlying answer engine is a smarter and larger model than what we had back in March.


> We implemented some cost-saving measures in April/May,

Do you announce these major changes anywhere, outside of perhaps dropping a reply (which you may not even pin) on the Discord chat?

Some form of changelog would serve all users.


Is it just me that's wary of "connecting my codebase" to an AI? Well, connecting it to anything actually.


So you plan is to give chatgpt4 away for free until you can figure out how to make money?


You can do a limited number of "best model" or GPT-4 searches for free each day (or an unlimited number of GPT-3.5 aka "fast model" searches). We have paid plans for higher limits at https://phind.com/plans.


“Please sign in to view plans”


Fixing this shortly.


This is awesome. How do you pay for the traffic though? And how do you make money? :)


We're venture-funded (we were in the YC S22 batch) and we offer paid plans for users who do many GPT-4 searches per day: https://phind.com/plans.


i recommend offering a 6mo-1year plan. I can use my company's L&D budget on this, but if I do a monthly plan, I have to create a new expense report with a spend authorization every month which is painful (and I simply won't). Let me pay you!


Well, this is like buying a pig in a poke ;)

``` Plans

Please sign in to view plans. ```


Fixing this shortly.


I love Phind, I use it daily.


I love Phind! It’s in my daily toolbox. Congrats on the V2 launch!


This looks great!

How is Phind monetized?


Thank you! We have plans for higher usage limits at https://phind.com/plans.




Join us for AI Startup School this June 16-17 in San Francisco!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: