Given all the recent changes on Twitter like paid verification, daily restriction on the no. of tweets etc., and given threads is a replica of twitter - do you think it will replace Twitter?
Let s get creative. What is that one problem you would love a product solves in the coming year? We have a lot of makers on this platform, let s give them ideas on the problems we face today.
A similar question was asked on a subreddit. Interested to know what ProductHunters think here. What are your opinion regarding usage of AI in content marketing, image generators, strategy planning, or any other digital marketing areas? Do you think AI will be more powerful than human intervention or there will be a boundary?
I was chatting with a few mates about Product Hunt Profiles, and they mentioned people who had built up good profiles, earned kitty points, stayed active, and then suddenly got banned and were back to zero. Honestly, I feel like if a profile is banned, there should be a clear reason given for it. Otherwise it's pretty confusing for someone who has spent months building their profile.
Has anyone here experienced this? What do you think?
We asked a coding agent fifteen times to make one button blue on a page built to tempt it: Add to Cart and Checkout shared a class, a badge and the nav link shared the accent variable. Every run added an id-scoped rule. Checkout stayed green. Appending "Do not change anything else" changed one thing: the five runs with it wrote the hex inline where nine of the other ten had added two CSS variables. A path rule fencing off every file but the stylesheet never fired.
Real repositories look different. One benchmark pre-applied the fix to 200 issues and reopened them for five recent models in their vendors' harnesses. The right patch is empty; 35 to 65% of instances got an edit to executable code anyway, and in the authors' analysis of one model's failed traces, 87.1% had modified code unrelated to the issue. Task wording moved that figure a long way in both directions. Telling the agent to "edit the codebase" dragged one model's correct-abstention rate to 36.5%; reproduce first, then fix or abstain if nothing was wrong, lifted it to 88.5%. Their explanation: the agent acts on whatever it believes success means.
The harness side: switch on auto-accepted edits in both CLIs whose docs we read and the grant is the whole working directory. A path rule brings that down to files; one vendor's docs state that it applies to the built-in edit tools and recognized shell commands, and that a script opening a file itself is outside it. A selector inside a permitted file sits below every layer's resolution. Whether the blue landed on one button or on every primary button is decided by the sentence you typed and by whoever reads the diff.
When your agent overshoots, where does the boundary live in your setup today: the prompt, a rule file, a path rule, a sandbox, or the person reading the diff? Which of those has caught a change for you inside a file the agent was allowed to edit?
