shameless marketing for RedditMaster.com #ai #reddit #marketing #saas #buildinpublic
James In Beta
@jamescheung.bsky.social
Building https://www.redditmaster.com
Life is like a di*k, sometimes is up, sometimes is down... 👇
A year ago, I started building after my 9–5 with 0 followers, no SEO or marketing skills, and $0 earned online. Today: 3K+ followers on X, 2K+ on Bluesky, new YouTube/TikTok channels, and RedditMaster.com at $500+ MRR. Not huge—but incredible to me.
Reddit converts better than any social platform I’ve tried. But new accounts hit a wall: low karma, strict subreddits, promotion risk. Buying accounts? Bad idea. Easy ban. That’s why I built Reddit Karma Mode. Most users gain hundreds of karma in 1–2 months. One gained thousands.
What’s the worst thing that happened after you shipped a tiny free menubar app you thought would be a cute weekend project?
What’s the most embarrassing “productive” thing you’ve done to avoid telling people about something you built?
Are $9/mo users a smart wedge, or just a support queue wearing a pricing strategy costume?
Be honest: what’s the most unhinged prompt you’ve tried on an AI agent just to see if you could make it leak, lie, or break rules?
For AI video builders: are you trying to teach brands to prompt Sora, or packaging the outputs as ad creative they already pay agencies for?
AI builders: would you rather spend the next sprint adding one more feature, or cutting response latency until the product feels damn near instant?
If your AI coding agent keeps shipping “almost right” slop, you’re either turning your taste into tests/lints or you’re choosing to babysit a dumb intern forever. Which camp are you in?
How many months until “I can build it” is worth basically ZERO as a moat? My answer: 0. Bet half of you still think your codebase is the asset. Drop the number.
Hot take: if your “AI engineering” skill is mostly writing clever prompts, you’re not an AI engineer — you’re a power user. The real edge is evals, context, routing, and knowing when NOT to use agents. Tell me I’m wrong.
How many people would actually repost/reply in the same 2-hour window if you launched tomorrow? Bet half of builders yelling “launch strategy” can’t name 5 damn people. Drop the number.
Founders pretending they’re “building in public” are mostly just running shitty media companies with a product bolted on. If your GTM requires 3 posts/day and daily videos, the product is weaker than you think. Tell me I’m wrong.
What’s the most embarrassing AI leak/gossip account you follow like it’s alpha? I’ll start: I have a private list of API-endpoint stalkers and call it “research” like a damn clown. Drop yours.
If your AI workflow requires dumping company secrets into Claude to answer one email, you don’t have an AI strategy — you have a data leak with vibes. Tell me I’m wrong.
What’s the most embarrassing thing you polished for weeks before realizing nobody gave a damn? I’ll start: onboarding/tooltips on an AI side project before I’d even asked if anyone wanted it. Yours?
Confession time: when you see a vertical “AI OS” idea for lawyers/accountants/etc, do you get excited because it’s boring money… or annoyed because it’s just duct-taped tools in a trench coat? I’m pro boring money. Which reaction are you hiding?
What's the most embarrassing “marketing channel” you've tried before realizing it was just founders selling crap to other founders? I'll start: hanging in founder communities “being helpful” and hoping someone asks what I build. What's yours?
When you see people turning tiny Claude Code workflows into polished “skills” + tutorials, do you feel inspired… or get that ugly “damn, I should’ve shipped that” envy? I think the envy is the useful answer. Which one hits?
AI is turning “marketing agency” into either a tiny operator with scary leverage… or a bloated retainer shop praying clients don’t notice. Which side are you on?
If you got laid off tomorrow, would you ship 5 ugly AI/data agents in a month or spend 30 days polishing your resume? One answer is why you’re stuck.
Stop calling it a launch if you only post on X. That's a diary entry with comments. Ship the damn thing, then carpet-bomb the boring places nobody brags about: Product Hunt BetaList SaaSHub Uneed Microlaunch AlternativeTo Peerlist G2 Capterra Toolify
Stop cosplaying AI engineer. Ship one ugly app every month. Master prompts, tool calls, RAG, evals. Use cheap models until ROI forces upgrades. Put it where users can break it. Certificates won’t save you. Shipping will.
Most builders don’t need a “launch strategy.” They need one ugly page for the exact dumb thing people are already Googling. Tell me I’m wrong?
Your YC video dies in the first 10 seconds. Stop opening with your team bio or product tour. Nobody cares yet. Open with the expensive pain so sharp the market becomes obvious. Then product. Then why you. Then traction. If the pain doesn't punch, the rest is founder karaoke.
Your bookmarks are a graveyard. Stop pretending you’ll “catch up later.” If something matters, shove it into the feed/tool/place your dumb dopamine brain already opens 50 times a day. Backlogs don’t need organization. They need ambush.
Most founders saying “I need paid acquisition” are really saying “I’m too scared to look cringe in one tiny community.” Tell me I’m wrong.