Fajar 🐬

built tools ai infra

[Codez @0xCodez]

https://t.co/eJB3TgyJwV

[rody @0x_rody]

https://t.co/LWQbDMXIdK

[Fajar 🐬 @fajarhide]

@gregisenberg @Rasmic @grok sebenarnya apa itu LOOP engineering yang di bahas disini?

[Akshay 🚀 @akshay_pachaar]

https://t.co/Exoyd8tB0d

[Hamza Khalid @humzaakhalid]

https://t.co/To5Vmqyli5

[AI_Explorer @ai_explorer25]

Best 15 accounts to follow in AI:

@karpathy = LLMs king
@steipete = built openclaw
@gregisenberg = startup ideas king
@rileybrown = vibecode king
@jackfriks = solo apps king
@levelsio = startups king
@marclou = startups king
@EXM7777 = AI ops + systems king
@eptwts = AI money twitter king
@godofprompt = prompt king
@vasuman = AI agents king
@AmirMushich = AI ads king
@0xROAS = AI UGCs king
@egeberkina = AI images king
@ai_explorer25 = AI queen

Follow them all and learn.

[Shubham Saboo @Saboo_Shubham_]

https://t.co/w7v8Vq1JNN

[Fajar 🐬 @fajarhide]

@karpathy Does vibe coding require an understanding of programming fundamentals? There is a lot of debate about this new term between experienced software engineers and new people who are just starting to create applications with vibe coding.

[Fajar 🐬 @fajarhide]

@Cointelegraph this would not have happened if they had known beforehand and used omni. because I saved about $35 this month after using it.

https://t.co/DpiQaiBBCD

[Vince 聊开发 @vincemask]

https://t.co/9LFFvTzNle

[Fajar 🐬 @fajarhide]

@vincemask Poin 2 and 5 covered by https://t.co/DpiQaiBBCD

try it!!

[Fajar 🐬 @fajarhide]

Github on https://t.co/DpiQaiBBCD

[Fajar 🐬 @fajarhide]

Everyone talks about better prompts.

Almost nobody talks about context management.

OMNI cuts 99.5% of terminal noise before it reaches your AI agent.

Docker logs: 9.2KB → 49 bytes
<100ms latency

Sometimes smarter AI simply means less noise.

https://t.co/DpiQaiBBCD

[Fajar 🐬 @fajarhide]

@namdoyan Dah coba ini bang? selain hemat dia juga selalu inget project yg lg lo kerjain, jadi gk recall2 context yg sama berkali-kali https://t.co/DpiQaiBBCD

[Fajar 🐬 @fajarhide]

Claude is smart, but debugging + long dev sessions keep hitting token limits 😅

Found a trick, use omni to control context.

Instead of Claude doing everything, Omni filters first → less noise, fewer tokens, better output. https://t.co/DpiQaiBBCD
@karpathy @claudeai @ClaudeDevs

[Fajar 🐬 @fajarhide]

@ojuschugh1 Nice share

can you visit my project too https://t.co/DpiQaiBBCD

[Fajar 🐬 @fajarhide]

Sekarang orang2 pada berlomba-lomba cari AI yg gratis dan yg hemat token, bukan lagi cari AI terbaik?

contohnya rtk muncul langsung viral dan akhir2 ini 9router muncul juga viral..

belum bisa bayangin, sebanyak apa nanti konsumsi token AI saat sudah ada AGI?

[Fajar 🐬 @fajarhide]

@indrazulfi Duren.. Selalu pengen makan tapi gak bisa sekaligus, harus di jadwalin makannya..

[Fajar 🐬 @fajarhide]

Approved contribute kecil gw ke project 9router, tapi jujur ini solve annoyance yg tiap hari gw rasain fiturnya. improve UX di “Add Model to Combo”, add model sekarang ada visual feedback, bisa inline toggle add/remove, tapi workflow jadi jauh lebih enak menurut saya.

[Fajar 🐬 @fajarhide]

Tools apa yg saat ini banyak di pake untuk hemat token AI?

[Fajar 🐬 @fajarhide]

@themishra4402 Nowadays, people are racing to find free AI or ones that cost less tokens.

For example, RTK went viral immediately, and recently, 9router also went viral.

How expensive will AI be in the future if AGI actually exists?