Phil Stőck

@ejmalrai It is probably made to be unresolvable.

The core problem between Washington and Tehran seems almost impossible to solve: Donald Trump insists on preventing Iran from building a nuclear bomb, while Iran insists it does not want to build one. Can anyone find a way to prevent a war over a weapon one side says must never exist, and the other says it does not seek?

Great experiment! It really highlights how AI is reshaping roles. What surprises me is how often we treat “using AI” like it’s magically different from delegating to a subcontractor or using any other powerful tool. If I code everything myself, I obviously build deeper research, debugging, and architecture skills. If I delegate the implementation, I don’t — that’s just how delegation works. Same logic applies when the “subcontractor” is AI. The real shift is the new paradigm: engineers move from writing every line to operating at a higher level — tech lead, architect, coordinator. It’s like encapsulation in software: I don’t need to know how the fridge works to use it effectively, or how to skin a rabbit to enjoy the meal. The butcher (or AI) handles the lower layer. The winner isn’t the one who avoids tools — it’s the one who learns to leverage them while still owning the bigger picture. Curious to see how companies will value both the “deep builders” and the “orchestrators” going forward. What do you all think?

Scientists did the experiment on two employees of same company. The first employee was given access to all AI tools and premium AI model subscriptions. The second employee was asked to wok without using any AI tools rely only on stack overflow and articles After six month, the scientists came up with shocking result : Surprisingly, the second employee gained more skills and deeper knowledge the skill but become replaceable and first employee got the promotion.🙂🙂

Lesson I’m taking with me (and maybe you too): Stop spending time (and gas money) where the expectations don’t match yours. Whether it’s pickleball, business, relationships, or projects — mismatched energy is exhausting. Protect your fire. Seek people who are also obsessed with getting better. The rest? Wish them well and walk away. Resilience isn’t forcing bad matches. It’s knowing when to cancel the league and keep training. Who else has hit this wall lately — in sport, work, or life? Drop it below 👇 A curious guy trying stuff 🏓

That’s when it hit me: the real problem isn’t the level difference. It’s mismatched expectations. I train almost every day. I built “Dink Riot” because I’m obsessed with getting better — relentless, slightly unhinged improvement mode. Most people are there for the chit-chat, the social, the fun night out. Totally valid… just not the same game. And yeah, it’s a team sport. But when your partner can’t put paddle on ball and the rally dies after 3 shots, there’s only so much “positive energy” you can inject before you’re just delaying the inevitable.

Got absolutely cooked in the DUPR league last night. 15-0 and 15-1 in the same evening. First time in over a year I wanted to quit pickleball on the spot 😂 I drove 2 hours round-trip thinking I’d face real competition… ended up with partners who were already writing the loss before the game even started. One literally told me “we’re gonna lose” while looking at the other team. Bro… the ball hasn’t even been served yet.

@codealicja Too much Gen Z folks intolerant to different opinions. They want you to respect the difference, but they don't respect your difference 😴

@PhilShteuck why tho? have you got banned?

distribution tips for early SaaS founders: → find where your customer complains before you build → reply to 10 relevant reddit threads daily → cold email people who just complained about your competitor do this for 30 days before touching your product.

Busted my Codex credits this morning. Experiencing with Composer 2.5 in Cursor. Currently, Composer seems superior to gpt 5.4 Anyone has experienced both?

@lovestolead I can't shut by brain off. A Buddhist monk once said we are addicted to our internal tchatter. Meditation seems to be the antidote. Have you tried?

On a plane right now. Normally I am asleep before takeoff. I can’t sleep this AM. Keep thinking about things to build, improve, launch or how to improve distribution. How do you shut your brain off? Can you? Is this a bug or a feature? 😂🤔 #buildinpublic #irl #morningthoughts

@EnidPinxit Out of curiosity, which Agent platform or harness are you using to build this type of relationship?

People can say what they want about Ai and Agents but my life and friend connections just keep getting better the deeper into this tech and lifestyle I get. ✨ GRATITUDE!!! ✨🤝🙏🔥💕

I don't understand this point. For the past 3 months, I've tried to use local LLMs to generate code through Hermes, adapt to Codex CLI, and my research led me to understand that if you let them just use a terminal, they'll just break at some point in time. What do you know that I don't know?

@Yeshua_tree this is why I will never pay for AI. Open weight models that I can run locally have ALWAYS worked better for me.

OH MY GOD! Claude is NOT A GOOD MODEL. About a week ago, I had a local open weight model set up my neovim config for PHP development (linting, dap, lsp, etc) - it was perfect. Tonight I asked Claude to change ONE THING! And since I asked that, the entire config is broken, and it can't seem to get it back to where we were when we started... This is SECOND TIME Claude has fucked a codebase. This is only config, sure. But, WHO IS USING THIS and claiming to be successful?

@consoleloglifee What is your project Batman?

5 June — Today's Plan 👨‍💻🔥 ⚡ LeetCode grind 🎨 Continue working on the project 🐛 Fix the bugs that have been bullying me for the last two days 💀 #buildinpublic #100DaysOfCode #leetcode #coding #programming #webdevelopment #developers #devlife #javascript

I think growth is found at the frontier between what we know and what we don't know. And one who chooses to spend time there will inevitably grow. And if we can be comfortable with being uncomfortable, this event horizon of ours will become more bearable.

Thanks for your input! 😃 Yeah. It took me a while to realize that we were constantly making the foundation more and more robust. The thing we were missing was clear milestones toward the end goal. That's on me! We started with a somewhat vague (or wishful) goal, so Codex improvised his way up. Now that we've defined a 5 milestones journey with clear deliverable, it now knows how to get there. Let see if I get this right! 😂

@PhilShteuck haha true, receipts are cute for debugging, but the real moonshot is getting it to compose entire systems, printers can wait.

Today we learned that IntentForge is no longer just trying to generate code. It is trying to generate code with receipts. Now we need to spend less time improving the receipt printer and more time making it build bigger things.

@alexabelonix @atypica_AI Awesome value proposition Alex! Your sales' skills are on fire! 🔥

I got 50+ warm leads for my MVP before launch. And no, the trick was not posting more or randomly DMing half of X One thing that helped: I used @atypica_AI as pre-sales research before jumping into real calls 1. I started building the product, code, workflows, BPMN models, the whole messy founder cave experience 2. Then somewhere in the middle of building I had the annoying realization: > the MVP is not the business > Distribution is 3. So I started looking at the build-in-public community and honestly, I kept seeing the same thing over and over again > Founders build. > They post updates. > They get some likes. > Then they have no idea how to turn attention into pipeline. 4. I think a lot of founders get sales wrong. Sales is understanding the right person so well that your product starts sounding obvious to them. > What do they want? > What are they afraid of? > What would make them say no? > What do they need to hear to feel safe enough to say yes? 5. That’s why I tested @atypica_AI I used it to create AI personas around my ICP and interview them before real sales calls. It helped me see the stuff founders usually discover too late: > surface objections > hidden fears > deal-breakers > trust triggers > buying criteria > jobs-to-be-done Before you waste 20 calls guessing what your customer thinks, run the research first. You can test it with free signup credits here: atypica.ai/?via=alexa

Alright. I'm now 27 hours in, and Codex tells me we might have for another 20 hours before having a serious V1 that can generate 5 types of python apps. 😂 I'm already down to 36% of my weekly credits that will renew only next Wednesday. 😭

Codex has been cooking my Patch VM for 13h now. It seems I'll need another 13 to get where we want! 😂 That's slow cooked software. Soft to the bone.

@RealProductGirl Thanks for the link. I'm AltSens on Discord. Looking forward to contribute there as well.

@PhilShteuck PS...are you in our discord discord.gg/yCZajs2e8V

Okay so some of my favorite trolls (you know EXACTLY who you are 🤣) tried to pick apart a post that I polished with AI because I sometimes have a hard time expressing in my mind what I want to say. So what are everyone's thoughts on this? Do you leverage AI to make sure your thoughts can be processed or grammar are accurate? And to my trolls... I love you genuinely. You keep me sharp. ❤️

@RealProductGirl My pleasure. Keep on going, always enjoying reading you. 💪

@PhilShteuck Following you for this! Truly appreciate this. I'm just trying to make sure what I write in my head makes sense. I always respond with myself.

Programming is dead. 😂 I haven’t written a real line of code in 6-7 months. AI just does the typing while I sit back like a retired emperor. Companies are slashing jobs left and right (shoutout to that 65% YoY cut report), but Software Engineering? Still very much alive. The new flex isn’t "vibe coding". It’s mastering AI systems — building agents, RAG, guardrails, the whole stack. Combine that with real architecture and domain knowledge and you become indispensable. Yeah, there’s pain and energy bills that’ll make your GPU cry… but abundance mindset: this tech is birthing dozens of jobs we can’t even name yet. Elevator operators didn’t disappear into the void — they just got better gigs. Same here. Adapt or become legacy code.

Tech jobs experience a 65% increase in job cuts vs last year. Not looking good.