HackerTrans
TopNewTrendsCommentsPastAskShowJobs

ActVen

no profile record

Submissions

Ask HN: Is any recent hourly social media activity data available?

6 points·by ActVen·2 anni fa·2 comments

Show HN: Create and Run a Team of Web Research AI Agents with a Prompt

app.soaringtitan.com
1 points·by ActVen·2 anni fa·1 comments

Comparing the responses of o1-preview and sonnet 3.5 to a team of agents

soaringtitan.com
2 points·by ActVen·2 anni fa·0 comments

Ask HN: Any tips on getting an invite to OpenAI Dev Day?

1 points·by ActVen·2 anni fa·0 comments

Creating a team of AI agents with a simple prompt

soaringtitan.com
1 points·by ActVen·2 anni fa·1 comments

comments

ActVen
·11 mesi fa·discuss
Just tried on Chrome instead of safari and it worked this time. Thanks and congrats on the launch!
ActVen
·11 mesi fa·discuss
Popup window with "Load Failed" after it had some progress on the bar past 40% or so. Shows up in the library, but won't play. I just deleted it for now.
ActVen
·anno scorso·discuss
I used to love using it. But, all of my coworkers hated it. I haven't revisited since the AI boom. Maybe it will have a bit of a comeback. Many people found it too difficult.
ActVen
·anno scorso·discuss
That caused a much needed chuckle this morning.
ActVen
·anno scorso·discuss
Way less than I used to. I have been a pretty advanced user since before google. The combination of AI and quick auto links to wikipedia articles on iOS have replaced much of it. The one place I still use it extensively is in local searches for businesses and when trying to find a brand or business that I know if they don't have an app.
ActVen
·anno scorso·discuss
Same question here. I have found repomix to get the job done really well.
ActVen
·anno scorso·discuss
Any unique benefits over using this vs something like Repomix? https://github.com/yamadashy/repomix
ActVen
·anno scorso·discuss
I really wish they would open up the reasoning effort toggle on o1 API. o1 Pro Mode is still the best overall model I have used for many complex tasks.
ActVen
·anno scorso·discuss
This is a great example of why people should not be afraid to be bold.
ActVen
·2 anni fa·discuss
Definitely. I'm just talking about an interesting way to identify content creation on a site.
ActVen
·2 anni fa·discuss
It might be very interesting to check your current traffic against recent api outages at OpenAI. I have always wondered how many bots we have out there in the wild acting like real humans online. If usage dips during these times, it might be enlightening. https://x.com/mbrowning/status/1872448705124864178
ActVen
·2 anni fa·discuss
Nice, that is a good idea. I think that result wouldn't be too surprising to anyone here. :)
ActVen
·2 anni fa·discuss
If anyone has any feedback on how useful the end product was or how it compares to other tools, I would love to hear it.
ActVen
·2 anni fa·discuss
The on device benchmarks are impressive. Especially tool use.
ActVen
·2 anni fa·discuss
We have been working on an internal tool that allows you to create a team of AI agents and tasks with a simple prompt. We find it very useful. If you have any feedback on the concept or might find something like this useful, please let us know. We are considering launching it as a project. Thank you!
ActVen
·2 anni fa·discuss
Here is the list of the questions. https://imgur.com/a/D4xwczU The PDF can't be shared. But, it looks something like the one here: https://content.naic.org/sites/default/files/call_materials/...
ActVen
·2 anni fa·discuss
Same here. Opus just crushed Gemini Pro and GPT4 on a pretty complex question I have asked all of them, including Claude 2. It involved taking a 43 page life insurance investment pdf and identifying various figures in it. No other model has gotten close. Except for Claude 3 sonnet, which just missed one question.
ActVen
·2 anni fa·discuss
Opus just crushed Gemini Pro and GPT4 on a pretty complex question I have asked all of them, including Claude 2. It involved taking a 43 page life insurance investment pdf and identifying various figures in it. No other model has gotten close. Except for Claude 3 sonnet, which just missed one question.
ActVen
·3 anni fa·discuss
The virtual objects and eye tracking are the features that, if they nailed it, seem to really set it apart. When they first launched the iPhone small details like the scrolling and pinch to zoom were just done so well, it changed the game with user interaction. I'm hoping the advances in this device have a similar impact on users.
ActVen
·3 anni fa·discuss
Has anyone seen a negative review from someone that has had a chance to try it a few times? All of the reviews I have seen have been pretty phenomenal. That is one of the reasons that I am looking to give it a try. If the UI and screen quality is so fantastic, it will be a great virtual screen for productivity and entertainment.