James Auble
804 posts

James Auble
@JamesAuble
Maker of https://t.co/YQx7QX0oZI & https://t.co/JzYHSiiVbE. Web developer. Freelancer.
California, United States Katılım Mart 2009
52 Takip Edilen43 Takipçiler

@psomkar1 It’s really easy. Purple gradient with a rocket ship logo. Bobs your uncle.
English

@suni_code I went for 48. Never looking back. Until 3 years from now.
English

@_brian_johnson I was gonna go 3 levels deep but I was trying to show restraint.
English

@JamesAuble vibe coding making a game about vibe coding is an elite level commitment to the bit
English

I vibe coded a game about vibe coding. You run a solo dev studio for 30 days, manage sanity and coffee, and watch a game build itself inside your game. Play it: vibe-coder.jamesauble.com
GIF
English

@Govindtwtt They're bad at making it pretty but great at making it work. Ship the ugly version, hire a designer later. Most startups die before they need animations.
English

@ayesha_fatiima Yes but not for the reason people think. You don't need to memorize syntax. You need to know when the AI is confidently wrong.
English

@Layton_Gott This is exactly how I built PayThread. Kept seeing freelancers complain about invoicing in every subreddit and forum. Built it, launched it, first users showed up without ads.
English

Step by step guide on how to validate your product idea...
The RIGHT way.
Step 1: Go where people complain. Reddit, X, App Store reviews, YouTube comments. Search your niche. If 10 different people in 10 different places are frustrated by the same thing, that's signal.
Step 2: Check if people already pay for bad solutions. Find competitors. If they have thousands of users and charge $20/month, the market is validated for you. Your job isn't to invent a market. It's to build something better.
Step 3: Build a simple MVP with a waitlist. Something that shows what the product does and lets people sign up for early access when the full version launches. Use AI to build it in hours, not weeks. You're collecting future users while proving the concept at the same time.
Step 4: Post it everywhere your target audience hangs out. X, Linkedin, Reddit, Indie Hackers, relevant Discord servers, communities. Share the problem you're solving and ask for feedback.
Step 5: Talk to the people who signed up. DM every single one. Ask what they're currently using, what they hate about it, and what they'd pay for a better solution. These conversations are worth more than any survey.
Step 6: Look at the data. If 50+ people joined the waitlist and 5+ said they'd pay, you have something. If nobody signed up and nobody replied, the idea needs work or the audience isn't there.
Most devs skip all of this and spend 6 months building and "marketing" something NOBODY wants.
Test the idea before you build the product. Always.
(Bookmark this for later)
English

@Niklas_Sikorra I keep hearing this while my SaaS keeps getting new users. Weird how the dead thing won't stop moving.
English

@kritikakodes It just means you got tired of context switching between two languages and chose violence.
English

















