It's simple. Drop your next big idea and get some brutally honest but hopefully valuable feedback. The rules are:
Drop your idea, you don't have to go super in depth but give a sentence or two about what it is and does.
Get roasted, duh
Roast someone else's.
That way we create a cycle of feedback
2K views
Replies
Best
Dropping this idea for roast:
A lightweight security scanner for Vibe Coders, Indie Makers who ship MVPs half-awake and only realize later we exposed half the internet.
Stuff it catches:
Exposed API keys, Misconfigured CORS, Missing CSP, CSRF, or secure headers, Public .env, .git, or test endpoints, JWTs with no expiry, Dumb stuff like buttons with no type inside forms, Laravel debug mode in prod, Vite dev server leaks, etc.
Gives a "ship risk score" + dead-simple fixes.
Not for security gods, it's just vibe coders or Indie Makers who don't want to get wrecked.
Would love brutally honest takes?
Report
@harvanshchaudhary security scanners already exist why would vibe coders need a different product?
@harvanshchaudhary@_tijs Just because scanners exist is not a reason not to make it. If anything, it means there's a market and people understand it. I love this because:
It's niched down by definition
It's easy to programmatically find stuff made in Lovable, Bolt, etc. so easy to do outreach
I say go for it.
Report
@_tijs Fair point! but most scanners are built for teams with infosec maturity. Vibe coders just want a "tell me what's wrong and how to fix it fast" vibe, not a 60-page OWASP checklist.
Think of it more like Clippy for code hygiene.
And funnily enough, this came from a real pain I saw in this Reddit thread. tons of devs shipping half-asleep and realizing later they had debug mode on in prod 😬
Report
@harvanshchaudhary yeah ease of use is the key. And the output could include prompts to fix each issue with whatever agent you have in cursor or zed
@harvanshchaudhary overall look & feel is clean, friendly. Broken on mobile, most people will see it on their phone so it should work there. Zip up your code and we’ll scan it is a non starter. That feels unsafe, it’s a hassle, you’ll have to re download the fixes? Meh. MVP should be a cursor extension or something along those lines.
Report
@harvanshchaudhary A security scanner for “vibe coders” who let AI tools like GPT churn out their MVPs while they’re half-asleep? Noble, but this is like putting a smoke detector in a house already on fire lol
Report
@kalebautomates Haha fair, the house is on fire, but at least I’m handing out fire extinguishers. Most of us are shipping half-baked MVPs anyway, so having a lightweight scanner to avoid total public humiliation feels like the least we can do before launch.
I'm developing DevBooster, a just-launched AI-powered browser extension that lets you grab any code snippet from the internet to instantly validate, explain, and improve it—perfect for self-learners and junior devs.
I'm open to any feedback—whether it's about the idea, the website, or the extension!
Report
Google for the subconscious mind
Imagine a startup founder hitting burnout. No time. No money for therapy. But he needs help.
He downloads Odyssea, googles his subconscious, and realizes it’s not burnout, but it’s “fear of failure”, coming up as the voice of his dad, who once failed in his bakery business.
He gets a tailored audio hypnosis, rewires his subconscious, and finally has the energy to show up for work.
Report
@tania_kodliuk This sounds so cool. I've been into meditating and rewiring my subconscious mind, and I've learnt that our manifestations and reality depend on our subconscious mind, and changing it isn't always the easiest for some. Your product is a game changer for people who get stuck in the loop of going back to old patterns of thinking or have blocks they haven't recognised as the stumbling block to a higher self. Just wow
Report
@ada_ubah Thank you so so much Ada! I feel you! I was on my journey too and it was very painful. The manifestation does not work if we have heavy emotions attached to the past. We need to understand and release emotions, learn our lessons, and then we can manifest anything. We have built Odyssea to help people get to understanding faster and in an easier way.
We do not need to have 5 toxic partners or bosses to learn the lesson of boundaries. Life can be easier.
Report
I want to make a chrome extension that allows the user to enter their work experience and a Job description, and the AI generates customized resume points for the user. I am still looking for ways on how I can build on this.
Report
@kushagrasinha I can suggest you ways to do this but if I understood correctly it is just a wrapper on an Existing LLM like chatgpt. Why you think this can be more useful than passing the same info to chatgpt asking for exactly what your application is claiming to return. Maybe I missed something from your description.
Report
@pankajti I'm planning on using a vector-database that consists of real resume bullet points, so the output is not very generalized and basic (Basic as in GPT will suggest similar structured points for every work experience a user enters). In simpler words, I want the bullets points to be more refined . I am still looking for ways on how I can turn this in to a better product and would appreciate any guidance. Thank you
Report
LifeLogs: Every day, the AI writes a 1-paragraph diary based on your calendar, photos, and texts. Your life, passively logged.
Report
@martianprompts sounds nice, but a bit too big brother-ish IMO. Definitely something Google could implement in near future.
Report
@scorbutics Totally get that, it does give off “Google-could-do-this” vibe.
Report
We're building a lightweight middleware library that requires minimal integration but delivers a significant boost to backend performance. It's designed to sit between your API and database, handling AI-enhanced prefetching and rule-based caching—perfect for apps dealing with complex data or performance bottlenecks. Drop it in, and your backend gets smarter without a full rewrite.
Would love to hear your thoughts or use cases where this could help!
Two-Step Contextual Enrichment (TSCE): first make an LLM spill a high-temperature, symbol-heavy anchor and then force another for the final answer to decode that anchor at T ≈ 0.1. On 300 GPT-3.5 prompts it jumped from 49 % → 79 % accuracy (+30 pp) and cut rule-breaks to 0/300, all for ~1.2 × token cost
Report
(Ok so I just discovered this community guys and that looks awesome!) Business Idea: A Social Network for Dance Enthusiasts
Objective:
Create a mobile application focused on social networking for dance lovers, featuring all dance-related events in the region and allowing users to share dance videos. The app would integrate editing tools, real-time annotations, and BPM-based suggestions for dance enthusiasts looking to improve their skills and discover dance events nearby.
Please, roast my idea before I spend too much time doing it lol.
Report
@scorbutics Very niche, but i'd say it might not work out only because there's tons of 'dance content' and communities already on ig and fb ;)
Replies
@harvanshchaudhary security scanners already exist why would vibe coders need a different product?
Product Hunt
@harvanshchaudhary @_tijs Just because scanners exist is not a reason not to make it. If anything, it means there's a market and people understand it. I love this because:
It's niched down by definition
It's easy to programmatically find stuff made in Lovable, Bolt, etc. so easy to do outreach
I say go for it.
@_tijs Fair point! but most scanners are built for teams with infosec maturity. Vibe coders just want a "tell me what's wrong and how to fix it fast" vibe, not a 60-page OWASP checklist.
Think of it more like Clippy for code hygiene.
And funnily enough, this came from a real pain I saw in this Reddit thread. tons of devs shipping half-asleep and realizing later they had debug mode on in prod 😬
@harvanshchaudhary yeah ease of use is the key. And the output could include prompts to fix each issue with whatever agent you have in cursor or zed
@harvanshchaudhary overall look & feel is clean, friendly. Broken on mobile, most people will see it on their phone so it should work there. Zip up your code and we’ll scan it is a non starter. That feels unsafe, it’s a hassle, you’ll have to re download the fixes? Meh. MVP should be a cursor extension or something along those lines.
@harvanshchaudhary A security scanner for “vibe coders” who let AI tools like GPT churn out their MVPs while they’re half-asleep? Noble, but this is like putting a smoke detector in a house already on fire lol
Visual Debug
I'm developing DevBooster, a just-launched AI-powered browser extension that lets you grab any code snippet from the internet to instantly validate, explain, and improve it—perfect for self-learners and junior devs.
You can check out the website and the extension at https://devbooster.ai.
I'm open to any feedback—whether it's about the idea, the website, or the extension!
Google for the subconscious mind
Imagine a startup founder hitting burnout. No time. No money for therapy. But he needs help.
He downloads Odyssea, googles his subconscious, and realizes it’s not burnout, but it’s “fear of failure”, coming up as the voice of his dad, who once failed in his bakery business.
He gets a tailored audio hypnosis, rewires his subconscious, and finally has the energy to show up for work.
@tania_kodliuk This sounds so cool. I've been into meditating and rewiring my subconscious mind, and I've learnt that our manifestations and reality depend on our subconscious mind, and changing it isn't always the easiest for some. Your product is a game changer for people who get stuck in the loop of going back to old patterns of thinking or have blocks they haven't recognised as the stumbling block to a higher self. Just wow
@ada_ubah Thank you so so much Ada! I feel you! I was on my journey too and it was very painful. The manifestation does not work if we have heavy emotions attached to the past. We need to understand and release emotions, learn our lessons, and then we can manifest anything. We have built Odyssea to help people get to understanding faster and in an easier way.
We do not need to have 5 toxic partners or bosses to learn the lesson of boundaries. Life can be easier.
I want to make a chrome extension that allows the user to enter their work experience and a Job description, and the AI generates customized resume points for the user. I am still looking for ways on how I can build on this.
@kushagrasinha I can suggest you ways to do this but if I understood correctly it is just a wrapper on an Existing LLM like chatgpt. Why you think this can be more useful than passing the same info to chatgpt asking for exactly what your application is claiming to return. Maybe I missed something from your description.
@pankajti I'm planning on using a vector-database that consists of real resume bullet points, so the output is not very generalized and basic (Basic as in GPT will suggest similar structured points for every work experience a user enters). In simpler words, I want the bullets points to be more refined . I am still looking for ways on how I can turn this in to a better product and would appreciate any guidance. Thank you
LifeLogs: Every day, the AI writes a 1-paragraph diary based on your calendar, photos, and texts. Your life, passively logged.
@martianprompts sounds nice, but a bit too big brother-ish IMO. Definitely something Google could implement in near future.
@scorbutics Totally get that, it does give off “Google-could-do-this” vibe.
We're building a lightweight middleware library that requires minimal integration but delivers a significant boost to backend performance. It's designed to sit between your API and database, handling AI-enhanced prefetching and rule-based caching—perfect for apps dealing with complex data or performance bottlenecks. Drop it in, and your backend gets smarter without a full rewrite.
Would love to hear your thoughts or use cases where this could help!
https://www.producthunt.com/p/self-promotion/turn-any-backend-into-a-speed-demon
A tool that utilizes AI to enhance video voiceovers. It enables users to upload any video and offers various options for improving the voiceover:
- Eliminate filler words and correct grammatical errors
- Recommend alternative scripts for better presentation
- Substitute the original voice with a selection of available voices
I feel it can benefit to those who want to create a demo video or an online lecture.
ProcessFlow - Turns your static process docs (Notion, Word, PDFs) into dynamic, branching workflows you can embed in tools your team already uses.
Here is the website:
https://www.process-flow.io/
https://github.com/AutomationOptimization/tsce_demo
Two-Step Contextual Enrichment (TSCE): first make an LLM spill a high-temperature, symbol-heavy anchor and then force another for the final answer to decode that anchor at T ≈ 0.1. On 300 GPT-3.5 prompts it jumped from 49 % → 79 % accuracy (+30 pp) and cut rule-breaks to 0/300, all for ~1.2 × token cost
(Ok so I just discovered this community guys and that looks awesome!)
Business Idea: A Social Network for Dance Enthusiasts
Objective:
Create a mobile application focused on social networking for dance lovers, featuring all dance-related events in the region and allowing users to share dance videos. The app would integrate editing tools, real-time annotations, and BPM-based suggestions for dance enthusiasts looking to improve their skills and discover dance events nearby.
Please, roast my idea before I spend too much time doing it lol.
@scorbutics Very niche, but i'd say it might not work out only because there's tons of 'dance content' and communities already on ig and fb ;)