... noticed a pattern in my own product work: When I'm deciding what to build next, it's very easy to make a feature list. "Add this." "Improve that." "Users might need this." So I tried changing the question. Instead of asking: "What feature should we build next?" I ask: "What assumption are we still uncertain about?" For example: If we think users will pay for a product, the uncertainty isn't "should we add Stripe?" It's: "Will someone ... ... build nothing and investigate the assumption first. I've found this especially useful while working on AI products, because AI makes it so cheap to turn an idea into something that looks finished. The harder part is deciding

... leadership." Copy, paste, post. Zero lived experience, zero original thought. It's just AI slop in a blazer. AI-written emails to people who matter. Your client can tell. Not because the grammar's off because the warmth is. VS GOOD AI JUDGEMENT Using AI to actually look at what a candidate did. Not keyword-matching a CV pulling apart their portfolio, cross-referencing their claims. You still decide

... than building an agent on top of it. 2. Pay yourself properly. There is already so much going wrong every day as a founder. You re earning far less than you did before, with none of the stability. I decided to pay myself the least on the team so we could put as much money as possible into the product and the people building it. That sounds honourable. In reality, you re adding more pressure to your life at exactly ... ... hear back about a $1 million investment. Imagine how crazy that felt. May We planned a three-week fundraising window and filled 75% of the round. Then a major UK VC offered to take the whole thing. We decided

... whatever tool can do it. Your product isn't the destination anymore. It's a capability the agent picks up mid-task, uses, and puts down. Nobody onboards. Nobody sees your landing page. The agent reads your tool definitions, decides if you're the right instrument for the step it's on, and calls you or doesn't. That reframes a bunch of maker questions at once. Your tool descriptions are now your marketing copy, and the audience ... ... model deciding

... book but not pay. somewhere, someone made that call. but how? was it intuition? a security review? a near miss? a customer asking a very uncomfortable question? because here's what i keep noticing: most teams haven't actually decided what their agent is not allowed to do. they've just decided

... start to lose my understanding of what it actually does. More features make sense only when the product is already established and wants to serve the needs of different types of users [cover a bigger market]. How do you decide how many features your product should have? (Are you a maximalist or a minimalist?) What helps you decide

... comes down to instinct built from watching the same three businesses long enough to recognize a pattern, not from anything a dashboard actually told me. Curious how other founders running more than one business, or more than one location, decide what gets handled first when everything looks equally urgent on the surface. Is it always the loudest voice, or is there something quieter you have learned to notice first ...

... case breaks the analogy exactly where you're pointing at it. The decider was never added to any directory of yours. There's no pool to provision them into ahead of time, because you don't know who they are until the moment you need them. So paging isn't a fix, it's a description of the easy case. The hard one, a genuinely external decider

... building a product around relationship-driven outreach, and one workflow keeps coming up: When someone has a goal like hiring, sales, fundraising, partnerships, or career exploration, the hard part is often not writing the message. It is deciding who is actually worth reaching first, why now, and what the smallest good ask should be. I am curious how other makers and operators handle this today. When you are trying to reach someone through your network: 1. Do you start with ... ... testing is: goal -> who to reach -> why now -> warm path -> first message If you work on sales, partnerships, hiring, fundraising, or career outreach, I would value your read on whether this matches how you actually decide

... works, payments in, landing page, guidelines followed. What I did not have was a single person who knew me before launch day. My Product Hunt account was four days old. Satchmo and Abdullah are right that the audience item decides the day. I would put it more bluntly: the other items decide

... businesses at once, and the hard part was never information, it's attention. Everything has a dashboard, but nothing tells you which fire is actually worth putting out today versus which one will burn itself out on its own. Curious how other multi-business folks decide

... worker. For UI testing, a remote headed browser is not just a convenience. It changes the debugging loop. Instead of asking "what did the script think happened?", you can ask "what did the user actually see?" ### Harness Code vs Local Agent Harness Another important decision is where the agent capability should come from. One option is to build the entire reasoning layer yourself: call an LLM API directly, manage prompts, retries, tool calls, screenshots, JSON parsing, and safety rules. That ... ... gives maximum control, but it also means the test harness becomes an AI platform project. The alternative

... Been thinking about this after a few good conversations here lately: there's a real tension between building something that flags "this looks off" for you automatically, versus just surfacing the raw signal and trusting the human to interpret it. Too much automation and you risk false confidence, the tool says everything's fine, so nobody looks closer. Too little, and you're back to relying entirely on someone's attention, which doesn't scale past a certain team size ... ... context. But I'm genuinely unsure that's the right call long-term. Curious what this community thinks: Do you prefer tools that flag problems for you, or ones that just show you the data and let you decide

... tarot better when it's human-led, not AI? Try a free card of the day and decide for yourself. Try it here: https://your-tarot.site I've been reading tarot for 15+ years. Most "tarot" sites today use AI-generated scripts that sound generic and hollow. That's not tarot. It's just text. So I built Mysticum a place where you get a real, human-led tarot reading. But I know not everyone trusts tarot (or even knows ...

... after tests already pass. What kind of behavior changes have surprised you most after accepting a refactor? Comment from Jason Scott(@jason_scott8): I have seen the same pattern where agents always suggest cleaning tested code. How do you decide when the readability gain is worth another careful review cycle? Comment from Omri Ben-Shoham(@omri_ben_shoham1): @Gal Dayan the blast radius framing is better than mine honestly, I was thinking about it purely in terms of "how ugly ...

... actually know how this work should be done, and how often will you run it? If you know the procedure and run it often, that's the part worth turning into a fixed, repeatable process instead of re-deciding by gut every time. Let the agent own what's genuinely unknown or one-off, and bound what you already understand. Chained autonomous steps compound errors fast, so the steps you can pin down are exactly the ones you should. Comment ... ... head right now which isn't great for consistency. Comment from King Joshua Marcos(@moddy): Honestly just trial and error early on. Let it fully own something low stakes, watched what broke, adjusted from there. The billing/auth vs

... launching Hello Inbox on PH as a B2B email deliverability SaaS. But, there's a back story I wanted to share with the community on how I validated the idea, and scaled to $12K MRR before deciding to build and launch a SaaS solution for my clients. I think this could be useful for startup founders who are trying to validate their business models. Because let's be honest, time is the most valuable resource. Sometimes the biggest lever ... ... deciding

... realizing I had spent the last 8 hours glued to back-to-back meetings without standing up, stretching, or drinking a single glass of water, it hit me: our calendars have become bullies. So, I decided to stop planning and start shipping. I built BreakMode . Think if Sunsama, Motion, and a Pomodoro timer having a baby. Except this baby doesn't force a complex workflow on you, and it doesn't charge you $30/month just to protect your right ...

... blurry, and I've talked myself out of good ideas by calling them "just one customer" when I should've listened. So where do you draw the line: how much of your roadmap do you actually let customers decide, and how do you keep that from turning into building whatever the loudest person asked for last? Comment from Farrukh Butt(@farrukh_butt1): I think customers should influence the roadmap, but not own it. A request is usually just a clue ...

... breaking, and so on. - Vaizle - I work with Vaizle, but it has very easily emerged as my fav platform. Not even lying. I can keep track of ad performance, check what competitors are posting on social. And best part is that I can feed it my own website + competitors

... thing I keep noticing is that almost everything can feel important when you are building. Posting content, talking to users, improving the product, replying to comments, checking analytics, fixing small bugs, testing new channels, networking, reading advice, watching competitors. All of it can be useful. But not all of it moves the business forward at the same stage. I m curious how other founders think about this. How do you decide

... they ask for. Focus on one thing, but keep testing channels. Build in public, but don t become distracted by content. I m curious how other founders filter this. When you hear advice that sounds reasonable, how do you decide whether it actually applies to your product or stage? Comment from Farrukh Butt(@farrukh_butt1): @evakk  That s a solid approach. I like the part about bringing repeated ideas back to the community, because it prevents feedback from becoming ... ... thats excellent advice because its backed up! Comment from Sherice Jacob(@kicksaascopy): The contradictions are real but most of it's because the advice is collapsed by stages, so people are compressing "What worked for me in the beginning" vs

... your question first, because you also handed me the variable that resolves it: it splits on reversibility, exactly as you guessed, but not only that. Two axes, not one. Reversibility decides whether real-time approval is even necessary, and blocking decides

... useful. Recently, I ve really liked concepts like @ProblemHunt and @404tomb , where people can validate an idea before building, whether there s actual demand for it, or whether that idea has already failed in the past. How do you decide if something is worth building? And how much time do you usually give an MVP before moving on? Comment from Sebastian Galvis(@klara_ai): From my perspective (a guy thats been in the VC industry and working with ... ... which was is there a painful workaround that people are already doing on their own? If someone is copying and pasting raw documents into ChatGPT and asking it to summarise them, that's a real signal. They've already decided

... Since sharing Prodshort here on @producthunt , we ve received a lot of early feedback. Ideas, feature requests, small improvements, UX feedback sometimes things we didn t even think about. How do you decide what to actually build from all that feedback? Some feedback looks useful at first, but once you test it, you realize it adds complexity without real value. Other times, a small suggestion turns into something essential for the product. Would love to hear from your experience ...