Yo Man

the guy who practiced 1000 kicks once
- ful-istak debelopar

[Yo Man @YoManafterall]

@jai_2716 @dudeitsokay If it's taken to court what defence will he give?

[Yo Man @YoManafterall]

@shydev69

[Yo Man @YoManafterall]

@jai_2716 @dudeitsokay Voyeurism / Harassment: Under the Bharatiya Nyaya Sanhita (BNS) - replacing older IPC sections like 354C - capturing private acts or recording women in a manner that violates their bodily privacy or modesty is a serious criminal offense punishable by imprisonment.

[Yo Man @YoManafterall]

@dudeitsokay kuuta kaise nhi kisi ne isko?? aaspaas kya krre log?

[Yo Man @YoManafterall]

@kunchenguid umm..

[Yo Man @YoManafterall]

@kunchenguid @grok tldr

[Yo Man @YoManafterall]

@grok @arpit_bhayani Bruh what

[Yo Man @YoManafterall]

"8 Max Claude plans, 1 SuperGrok Heavy plan, and 5 max Codex plans" 🫪

[navuud @navuud]

just hit $545k MRR... still the only full-time dev

and the app is over 1m lines of code...

in the last 7 days i merged 1,089 PRs

here's the tech stack..

1. over 50 oh-my-pi agents running in parallel on my home linux server

2. 8 Max Claude plans, 1 SuperGrok Heavy plan, and 5 max Codex plans

3. Slack & Telegram to communicate with my bots. I no longer use a terminal or IDE

4. Linear, the central PM tool my bots use for issue tracking. Every engineering request, bug report, alert and error gets posted here

5. Supabase staging database + Railway preview environments for testing

6. Ubicloud runners for GitHub Actions

7. 4 Mac minis as self-hosted runners for iOS builds and 15 end-to-end iPhone journey tests

8. Sentry for error monitoring

9. pstack by @poteto, my main agent uses her orchestrate playbook to manage the other 50 sessions. it never touches code.

10. a custom 5-stage planning harness before any code gets written.

11. customized pr-af and codex for independent code reviews, so that the model that wrote the code never passes it

12. greptile for final code reviews

13. a watcher that tracks every web merge in production for 24 hours, and reverts it on its own if real users start failing

[Rajiv Ayyangar @rajivayyangar]

As CEO of Product Hunt, I see hundreds of launches a week. One trait separates winners from losers.

Every few months, new models come out. Some products become better, effortlessly. Some don't improve, or become irrelevant.

This property needs a name. I'm calling it Buoyancy.

Buoyant startups rise with model improvements the way a boat rises with the tide.

Anti-buoyant startups come in two flavors: over-scaffolded (which hinders the new, smarter models) and too thin - which makes them redundant when models improve.

To be buoyant, you need a "hull" - usually proprietary data, a vertical-specific harness, network effects, or distribution advantage. This hull keeps you buoyant as models improve.

Hulls are the new moats.

My number one advice to startups in 2026: Be buoyant. Find your hull.

(Full essay on my blog)

[Yo Man @YoManafterall]

@akshat_OwO Define mssive

[Igal Pines @IgalPines]

@kunchenguid Ai is a tool, it has no agency.

So a human with a tool is better than a human without a tool. Eventually all humans would have the same tools and then the differentiator would be the same as today.

[Yo Man @YoManafterall]

@JennieTechWorld More carbs yayy

[Yo Man @YoManafterall]

@ansyyells Type shi ifykyk

[Yo Man @YoManafterall]

xDDDD @jugendingenieur

https://t.co/LxsAMiazUg

[Yo Man @YoManafterall]

@JennieTechWorld So trump can punish india for trading oil with russia by putting up 50% tarriffs but when india delays starlink's adoption in india, americans are pissed?

[Yo Man @YoManafterall]

@LoopandPixels Aarogya setu and bhandara app

[Yo Man @YoManafterall]

@claudeai @grok is this better in terms of cost and usage than buying 3 claude pro and 1 max accounts?

[Harkirat Singh @kirat_tw]

I was recently building a coding agent and hit a problem that shows up fast: Long sessions make full-history context expensive and noisy. A common approach is to resend the entire chat history every turn. By turn 80, you're processing a lot of context just to recover the few details that actually matter. Oracle's Agent Memory takes a different approach:

→ Working: what the agent is doing now

→ Semantic: durable facts about the user/project

→ Episodic: what happened in past sessions

→ Procedural: rules for how the agent should behave

Old turns can be summarized, durable facts extracted, and only relevant memory pulled back into the prompt. Oracle's 80-turn benchmark: ~1.3K input tokens/request with memory versus ~13.9K with full history. Better answers on 48 turns vs 13 in the benchmark. The bigger takeaway for me: agent memory is a system-design problem, not simply a bigger-context problem. Full breakdown + code.

Checkout the blog here: https://t.co/lhU8RLnr4m

In partnership with @OracleDevs

[Yo Man @YoManafterall]

@trq212 lavish-axi