himiespa_untad

Himpunan Mahasiswa Ilmu Ekonomi dan Studi Pembangunan.
Sekretariat : Jl. Tomblotutu

[Juliette Delorme @DelormeJul11265]

🇦🇺 Un homme. Un drapeau. Une foule pro-génocide à Bondi Beach.
Il s'est tenu seul, a levé le drapeau palestinien et a fait taire des milliers de personnes.
Pas d'armée, pas d'armes—juste un morceau de tissu qui a effacé tous les autres symboles. Voilà ce qu'est le courage. ✊🇵🇸

[himiespa_untad @himiespa]

Some days are meant for adventure, others for slowing down

https://t.co/2ou5iTr3Ln

#camping #campingtents #inflatabletents #popuptents #tentstore #tentshop

[BuBBliK @k1rallik]

AI APPS DON'T HAVE TO LOOK LIKE AI APPS

Same gradients. Same rounded cards. Same forgettable design.

Bolt just launched a Template Marketplace built to break that pattern.

- Real full-stack apps with the backend and logic already working
- Designs created with some of the world's best studios and agencies
- Everything can be customized with prompts
- Every template is free to open and use

Stop prompting from zero. Start from something worth shipping.

[kompot @ethkompot]

A billionaire publisher sat down with a fund manager and admitted on camera that the boards he sits on fire the managers who are about to do well.

The men are Joel Greenblatt and Steve Forbes. Greenblatt ran Gotham Capital from 1985 to 1994 at roughly 50% a year, turning 7 million dollars into 500 million, then returned all outside capital and started teaching at Columbia.

An armchair, a table lamp, bookshelves, a glass of water on the floor. April 2010, part 2 of a conversation filmed in somebody's sitting room.

Forbes says he sits on endowment boards, and describes what happens in those rooms when a manager has 3 bad years. Greenblatt takes it from there.

The pattern is mechanical. Committees hire on trailing performance and fire on trailing performance, which means the money arrives after the good run and leaves before the next one.

The turn is that this is not a story about incompetent committees. It is the structural reason a published strategy keeps working. Somebody has to be selling at the bottom, and institutions are organised so that it is them.

It matters more now. Every allocator has faster reporting, shorter review cycles, and the same 3 year patience they had in 2010.

Free on YouTube, 10 minutes, one fixed camera, part 2 of a conversation nobody finished.

The formula was public. The committee structure is the moat.

3 bad years. 1 sitting room. It is in the video.

[Javier Negre @javiernegre10]

Caos en Japón.

Miles de japoneses se reunieron para protestar contra la construcción de la primera mezquita en Fujisawa.

La mezquita propuesta sería más grande que el santuario sintoísta, además quieren instalar altavoces para el rezo 5 veces al día, un acto de provocación.

«¡No queremos ni una sola mezquita, ni cementerio musulmán aquí!» gritan.

[himiespa_untad @himiespa]

Some days are meant for adventure, others for slowing down

https://t.co/2ou5iTr3Ln

#camping #campingtents #inflatabletents #popuptents #tentstore #tentshop

[KPK @KPK_RI]

Jakarta, 31 Juli 2026.

KPK menahan empat orang tersangka dalam perkara dugaan tindak pidana korupsi terkait pembayaran komisi terhadap asuransi perkapalan milik PT Pelni oleh PT Jasindo (Persero) tahun 2015-2020 yang telah menyebabkan kerugian negara.

[himiespa_untad @himiespa]

Some days are meant for adventure, others for slowing down

https://t.co/2ou5iTr3Ln

#camping #campingtents #inflatabletents #popuptents #tentstore #tentshop

[🇨🇳XuZhenqing徐祯卿 @XueJia24682]

✨🇨🇳 BYD is constructing its global R&D center in China.
The 20-billion-yuan project will become the world’s largest NEV research hub.

[Universitas Tadulako @univ_tadulako]

Rektor Tutup PKKMB Tingkat Universitas, Tekankan Pentingnya Karakter, Integritas, dan Literasi Digital Mahasiswa https://t.co/tOqIsf0Xsq

[BBC News Indonesia @BBCIndonesia]

Tiga warga sipil ditembak milisi TPNPB atas tuduhan jadi mata-mata militer, 20 Juli lalu. Sepanjang sejarah konflik di Indonesia, dari Aceh, Timor Timur, dan Papua, perang gerilya selalu "membuat warga terpojok". Apa solusi yang harus diambil pemerintah? https://t.co/VLi7Cq3zcK

[himiespa_untad @himiespa]

Some days are meant for adventure, others for slowing down

https://t.co/2ou5iTr3Ln

#camping #campingtents #inflatabletents #popuptents #tentstore #tentshop

[himiespa_untad @himiespa]

Some days are meant for adventure, others for slowing down

https://t.co/2ou5iTr3Ln

#camping #campingtents #inflatabletents #popuptents #tentstore #tentshop

[scalp @0x_scalp]

YOUR NOTE APP IS A GRAVEYARD. THIS TURNS IT INTO AN ATM.

400 notes. Zero connections. Claude reads all of them in one session.

Drop a source → Claude ingests it, pulls out every idea and person → cross-links it to everything already in the vault → files it into clean Markdown you own.

Ask it anything 3 months later. It answers in 2 seconds and cites the exact pages.

The vault structure that prints:

→ raw/ — drop sources here, Claude reads, never edits
→ wiki/ — Claude writes here, 8–15 linked pages per source
→ output/ — finished work, slide decks, query results
→ one CLAUDE.md — your profile, loaded every session, you never re-explain yourself again

95% of people miss the real unlock: point every Claude Code project at the same vault. One brain. Every project drinks from it.

Knowledge compounds like interest. Most people let it rot in tabs.

The creator in this video builds and sells this setup to clients. The graph you're looking at isn't decoration — it's the product. And it runs for $0 on top of Claude you're already paying for.

2 lines to clone it. 1 word to scaffold it. Full build in the article 👇

[rmen @Roman9078963816]

EVERYONE IS TALKING ABOUT BETTER AI MODELS.

Almost nobody is talking about who owns the compute.

That's where the real shift is happening.

Most developers still pay every month to run AI.

Cloud GPUs.

API credits.

Usage limits.

Recurring bills that never stop.

Some are choosing a different path.

Instead of renting compute...

They're buying it once.

One example is the NVIDIA DGX Spark.

What you get:

→ 128GB unified memory

→ 4TB SSD

→ Run large open-weight models locally

→ Private AI workflows

→ Your data never leaves your machine

No hourly GPU pricing.

No API rate limits.

No waiting in cloud queues.

The same box can power:

• AI agents

• Private RAG

• Coding assistants

• Research pipelines

• Enterprise knowledge bases

• Local automation running 24/7

The biggest change isn't a faster model.

It's turning AI from a monthly subscription...

...into infrastructure you own.

Cloud made AI accessible.

Local hardware is making it independent.

Five years from now, we may look at renting GPUs the same way we look at renting servers for every line of code.

The AI race is no longer just about choosing the best model.

It's about deciding who owns the computer running it.

Bookmark this. The AI PC era is only getting started.

[himiespa_untad @himiespa]

Some days are meant for adventure, others for slowing down

https://t.co/2ou5iTqvVP

#camping #campingtents #inflatabletents #popuptents #tentstore #tentshop

[Misato @misat0x]

A YouTube channel with fewer than 25,000 subscribers made $34,000 in 90 days.

But...

the part people will copy from this video is probably the least valuable.

Claude finds recently launched channels, profitable niches, and estimated revenue.

Useful.

But a table of high-RPM niches is not a YouTube business.

Most people enter YouTube the wrong way around.

→ They pick a niche.
→ Spend days making a video.
→ Then discover nobody wanted it.

AI makes that loop much cheaper.

→ Find what is already getting watched.
→ See where the demand is.
→ Study the angle competitors missed.
→ Then use AI to turn one validated topic into a sharper script, faster edit, and better thumbnail.

The part I keep coming back to:

AI can make production cheap.

It cannot make people care about your version.

That still comes down to taste, a real point of view, and knowing when a video is boring before the audience tells you.

The creators who win will not be the ones generating the most videos.

They will be the ones who point AI at real demand before everyone else does.

Full article below👇

[himiespa_untad @himiespa]

Some days are meant for adventure, others for slowing down

https://t.co/2ou5iTqvVP

#camping #campingtents #inflatabletents #popuptents #tentstore #tentshop

[Mark Poiler @mpoilerfx]

A 27-year-old machinist built a single-tower local AI rig on his kitchen table and now runs 27B models at 30+ tokens/sec for pennies per hour.

No cloud bills. No rate limits. Full privacy.

He spent $1,650 on the case, board, cooler, and 128GB of RAM. Power draw sits near 120W at load. His electric bill moved $11 last month.

The wins: Local Qwen3.6-27B scoring 77.2% on SWE-bench. Private document Q&A over his own files. Agents grinding overnight while he sleeps. Zero vendor lock-in.

The pain: Feeding it 50-page documents tests his patience, prompt processing crawls next to NVIDIA's box. Big fine-tuning stays out of reach. The hardest reasoning still goes to a cloud API. And the RAM he paid $400 for now sells near $700 after DRAM prices jumped 90% in Q1.

He still runs it every day. The box answers before a browser tab would load, and nothing leaves his network.

One quiet tower glowing on the table. The fans barely whisper. The meter barely moves.

Worth it.

[himiespa_untad @himiespa]

Some days are meant for adventure, others for slowing down

https://t.co/2ou5iTr3Ln

#camping #campingtents #inflatabletents #popuptents #tentstore #tentshop