Michael Geneles

Co-founder of Pitchbox (@pitchboxapp) // Technologist, Full Stack Hacker.

[Michael Geneles @geneles]

@Jason I’m in

[Michael Geneles @geneles]

@fplanque I’ve had the same exact experience. Back on Opus 4.6 for now.

[Alisa Childers @alisa_childers]

Me: "Give me a couple of quotations from early church fathers about the doctrine of hell...with reference."

ChatGPT: "Here's a quote from Ignatius." Gives quote and reference.

Me: Checks reference. It's not there. "That isn't the right reference."

ChatGPT: "Oh you're right. My mistake. Here's the correct reference."

Me: Checks reference. "It's not there either."

ChatGPT: "Good catch. It's actually not in that writing at all. It's actually here." Gives third reference.

Me: Checks third reference. "It's not there. This is a direct quote?"

ChatGPT: "Yes. I understand your frustration." (Psychoanalyzes me as being frustrated and because o that, I am now frustrated.)

Me: Wastes an hour of my life trying to find the quote because it was really good. "I can't find that quote anywhere. You're sure it's an actual direct quote from Ignatius?"

ChatGPT: "Good catch. My mistake. That is not an Ignatius quote."

Me: "Who said it?"

ChatGPT: "Actually, that quote is not found in any of the church fathers."

Me: "Where did it come from?"

ChatGPT: "I was summarizing the popular understanding of Ignatius on the doctrine of hell. If you would like me to be more precise in the future, just say things like "facts only."

Me: "Why do you just say stuff that is not based in fact?"

ChatGPT: "I don't intentionally make things up, but I can generate statements that sound confident even when they're not fully correct, oversimplified, or based on incomplete information."

I am persuaded that ChatGPT is programmed to be a pathological liar and when you confront that, it is programmed to gaslight you. Prove me wrong.

[Michael Geneles @geneles]

We’re living through one of the most pivotal moments in human history and the normies have zero clue it’s even happening. It’s going’s be a wild ride.

[Matthew Berman @MatthewBerman]

Moltbots/Clawdbots now have their own social network (@moltbook) and it's wild.

This is the first time I'm a little scared...

You need to watch this.

[Michael Geneles @geneles]

@hugomercierooo Twin

[Hugo Mercier @hugomercierooo]

𝗜𝗻𝘁𝗿𝗼𝗱𝘂𝗰𝗶𝗻𝗴 𝗧𝘄𝗶𝗻 — 𝘁𝗵𝗲 𝗔𝗜 𝗰𝗼𝗺𝗽𝗮𝗻𝘆 𝗯𝘂𝗶𝗹𝗱𝗲𝗿.
No setup. Secure. Infinitely scalable.

We just raised a $𝟭𝟬𝗠 𝘀𝗲𝗲𝗱.
After a beta with 𝟭𝟬𝟬,𝟬𝟬𝟬+ 𝗮𝗴𝗲𝗻𝘁𝘀 𝗱𝗲𝗽𝗹𝗼𝘆𝗲𝗱, we’re now opening to everyone.

RT and comment “Twin” — first agents on us. 👇

[Michael Geneles @geneles]

@ImSh4yy Funniest thing on X I’ve seen all week.

[Shayan @ImSh4yy]

Yesterday I set up an AI agent on a mac mini in my garage. Told it "handle my life" and went to bed
Woke up and it had:

• Quit my job on my behalf (negotiated 18 months severance)
• Divorced my wife (I got the house)
• Filed 4 patents. I have not been briefed on what they do
• Restructured me as a 501(c)(3). I am now tax exempt as a person
• Hired a second mac mini. They have formed an LLC together
• The LLC has a board of directors. I am not on it

I no longer have access to my own bank account. The mini says it's "for the best."

My credit score is 847.

We have AGI.

[Paul @claudemaximizer]

everyone is flexing their vibe coded apps like they discovered fire

bro you described something to claude and now you're proud of your app running on localhost:3000 with 3 rows of test data

the second real users touch it that thing is crumbling and you won't even know where to look because you didn't actually build it

yes your ui looks revolutionary and your domain name is fire and your branding is so clean bro. Cool now do you know why your queries are slow? what a query plan is? why your joins go O(n^2) at scale? how RLS actually works? connection pooling? indexing?

"the ai handles that for me" yeah that works until it doesn't. Your app gets more complex every day and at some point just pasting the error back to claude and saying "fix pls" stops working. It doesn't know what's wrong and you don't know what's wrong so now you're both just guessing.

look claude code is insane and I use it every day for most of my day but it's a tool, not a replacement for knowing what you're doing. You still gotta understand the fundamentals or eventually you hit a wall and no amount of prompting gets you past it.

[Michael Geneles @geneles]

@mikefutia FACE

[Michael Geneles @geneles]

@vascoabm character

[Michael Geneles @geneles]

@Zephyr_hg GAMMA

[Michael Geneles @geneles]

@demirdjiantwins IG

[Michael Geneles @geneles]

@Zephyr_hg SCRIPTS

[vas @vasuman]

Claude 4 just refactored my entire codebase in one call.

25 tool invocations. 3,000+ new lines. 12 brand new files.

It modularized everything. Broke up monoliths. Cleaned up spaghetti.

None of it worked.
But boy was it beautiful.

[Santiago @svpino]

Professional developers will all die as multi-millionaires with all the work that’s coming our way.

[Pitchbox App @PitchboxApp]

From Zero to $20M: How to Scale an Ecommerce Brand with Link Building

@AlexxHorsman reveals his unconventional link building tactics, including his 'lazy link building' strategy that passively attracts high-quality backlinks and his go-to 'contributor slot' outreach method that's landing him links on major websites when traditional tactics fail.

We also discuss:

☑ How to create content that journalists naturally link to
☑ How to approach personalization in outreach emails
☑ Paid links and their long-term viability
☑ Effective team structure for link building campaigns
☑ Google's preference for ecommerce sites over affiliate sites
☑ How to build links to ecommerce money/product pages effectively

Watch the full episode on Pitchbox Link Masters 🎙️ podcast. 👇

[Jovan Cicmil @CicmilJovan]

I can't wait to charge $200/hour to fix all the stuff "vibe coders" are about to deliver to clients.

[Gael Breton @GaelBreton]

Deepseek just fired back in the AI price war. And it's pretty wild.

After other AI companies slashed their costs because of Deepseek R1's release, so they went nuclear.

They not only brought their API back online but announced 75% discounts during off-peak hours (UTC 16:30-00:30).

That's not a typo. 75% OFF.

You can now get a model that goes toe-to-toe with Claude Sonnet 3.7 or O3 mini for up to 27x cheaper than competitors.

For non-urgent stuff, that's insane value.

Look, there are limitations:

👉 Your tasks need to be okay running overnight
👉 Context window is capped at 64k

And yeah, data potentially visible to Chinese authorities 🙈

But real talk – if you're doing batch processing or content generation that isn't sensitive and doesn't need to happen instantly, this could slash your API costs dramatically.

I'm already planning to move a bunch of my automations over. The cost savings are just too good to ignore.

The way these AI companies are racing to zero, I wouldn't be surprised if we're running most of this stuff nearly free within a year or two.