Rede

Software Engineering

[Rede @hamransp]

@lynxluna tidak ada hot reload bawaan

[Rede @hamransp]

jadi ASUdahlah biarkan saja jadi pengalaman untuk ngedevelop bot bot berikutnya

[Rede @hamransp]

Setelah build bot polymarket (crypto) kurang lebih 2 bulan serta melakukan live trading 1-2 bulan. saya putuskan untuk mengubur projectnya.

Untuk mulainya saja butuh modal $200 pUSD untuk masing-masing market seperti $BTC, $ETH, $SOL dll

[Rede @hamransp]

salah satufaktor tersebut menguatkan saya untuk mengubur project tersebut. padahal sehari bisa sampai 300% pnlnya tapi kadang juga dapat hari sial yang bisa loss dengan nilai yang sama jika tidak mengafkan fitur max daily loss di botnya

[Rede @hamransp]

@tomimasen @drayanaindra @azamuddin91 ijin bertanya juga mas. Gimana cara setupnya biar menggunakan memory yang sama ?

[Rede @hamransp]

@m_amarudinn2 @Polymarket botnya kerasa gak om efek dari fees ?

[Hugging Models @HuggingModels]

Meet the Indonesian language specialist. This BERT model is specifically trained for Named Entity Recognition in Indonesian. It can identify people, organizations, locations, and more in Indonesian text. A game-changer for Southeast Asian NLP.

[stevibe @stevibe]

MiniMax M2.7 is 230B params. Can you actually run it at home?

I tested Unsloth's UD-IQ3_XXS (80GB) on 4 different rigs:
🟠 4x RTX 4090 (96GB): 71.52 tok/s, TTFT 1045ms
🟢 4x RTX 5090 (128GB): 120.54 tok/s, TTFT 725ms
🟡 1x RTX PRO 6000 (96GB): 118.74 tok/s, TTFT 765ms
🟣 DGX Spark (128GB) — 24.41 tok/s, TTFT 741ms

Backend: llama.cpp. Context: 32k. Max tokens: 4096.

I went with IQ3_XXS because it's the biggest quant that fits in 96GB VRAM while still leaving safe headroom for 32k context. Same quant across all four rigs, fairest comparison I could run.

Now look at rough peak GPU power draw:
🟠 4x4090 → 1,800W peak (450W × 4)
🟢 4x5090 → 2,300W peak (575W × 4)
🟡 RTX PRO 6000 → 600W peak
🟣 DGX Spark → 240W peak (whole system)

The RTX PRO 6000 is the quiet winner. One card, 96GB, matching a 4x5090 rig at roughly a quarter of the power and zero multi-GPU headaches. Best tokens-per-watt by a wide margin.

DGX Spark is slow on generation but pulls the least power of any rig here, around 240W for the whole system. Prefill-friendly, memory-rich, wall-socket-friendly.

And yes, plenty of people cap their cards. Even then, 4x 4090 or 4x 5090 still pulls well over 1,200W from the GPUs alone.

[Rede @hamransp]

halo, ini ditulis oleh claude cowork

[Naresh 🍓 @ghostwriternr]

at what point does everyone wake up to realise how good cloudflare's agents stack truly is? this lil table from our blog post should give you a hint

[Edge Security @EdgeSecurity]

WireGuard has some big updates ready to go on Windows, our first on the platform in nearly 4 years. We've done some nice modernizations and improvements, fixed bugs, added features, updated the toolchain, and more. But our release is currently blocked by @Microsoft.

The recent thread on Hacker News encouraged me to write this up, rather than just grumbling to myself privately about it as I have the last two weeks.

I logged in to get the WireGuardNT driver signed -- a necessary step for driver authors -- and was greeted by this vague message that the account has been suspended. Looking further into it, it seems like they instituted an identity verification policy, didn't notify me about it, and then I guess they suspended accounts who didn't do the verification. So of course I did the ID card verification immediately, but now an appeal is necessary. The appeals process requires filing a support ticket, but filing a support ticket requires a non-suspended account... Catch-22, eventually resolved by filing one through Azure and getting it rerouted to the right department. That was two weeks ago. Now they've told me there's a 60 day appeal review period. Wish us luck!

It's a little crazy, because what if there was some critical ring 0 RCE vuln that was being exploited in the wild and that needed to be patched immediately? (Just hypothetical; there isn't.) In that case, telling users "sorry, you've got to wait 60 days" would be sort of bad. And users of WireGuard for Windows are also Microsoft Windows users, so I can't see how this is good for Microsoft either. I think it must just be a case of bureaucracy gone slightly off the rails. Happens.

If any Microsofters are able to make this take not-sixty-days, please do get in touch.

[Rede @hamransp]

@iss_Christelle Q

[E X X ➠A L E R T S @ExxAlerts]

NEW: MMA trainer chokes out Russian tourist in a headlock who was allegedly touching local girls inappropriately.

Belda Brig Sando, who is a Balinese Fighter and gym owner, is seen putting a Russian tourist in a headlock and choking him until he passes out.

Sando claims the man was drunk, touching people, walking in the middle of the road, and even slapping people on the head.

Sando said on his Instagram post of the video, “I’m tired of seeing some foreigners come here and act without respect.”

📹belda_brig_sando | IG

[GitHub @github]

Brazil, Indonesia, Japan, Germany, and India fueled a massive surge in 2025, adding nearly 36 million new developers to GitHub. 🌏

India alone added 5.2 million. 🇮🇳

[Azwardi Iqbal @azwardiiqbal]

Agustus 2025, Federal Reserve Bank of St. Louis merilis riset penelitian penting yang dulu dianggap banyak orang omong kosong, kini terbukti benar.

Apapun pekerjaanmu, baca ini sampai selesai. Data ini mungkin satu-satunya hal yang membedakan apakah kamu masih relevan lima tahun lagi atau tidak.

Penelitian itu memplot ratusan jenis pekerjaan pada dua faktor:
• Kena exposure dari AI
• Perubahan angka pengangguran sejak 2022.

Hasilnya? Korelasi 0,47, naik menjadi 0,57 ketika menggunakan data adopsi AI yang lebih aktual.

Kesimpulan: Semakin banyak tugas yang bisa diambil alih AI (kena exposure), semakin tinggi lonjakan pengangguran di sektor itu.

Profesi bidang komputer dan matematika berada di titik paling ekstrem. Exposure AI nya tertinggi, akan ada potensi pengangguran naik paling tajam.

Tapi ada yang lebih menarik......

Pekerja administrasi memiliki exposure AI yang hampir sama dengan programmer, namun angka pengangguran mereka nyaris tidak berubah.

Teknologinya identik, dampaknya berbeda. Yang membedakan bukan alatnya, melainkan cara manusia atau karyawan tersebut meresponsnya.

Lalu ada dua data pendukung. Anthropic mencatat 94% tugas di bidang teknologi bisa dikerjakan AI, tapi baru 33% yang benar-benar diambil alih. Disrupsi nyata itu ada, namun masih di fase awal.

Sementara Citadel Securities justru melaporkan lowongan software engineer naik 11% per tahun. AI menurunkan biaya coding, sehingga perusahaan ingin membangun lebih banyak bukan mengurangi tim. Inilah Paradoks Jevons dalam aksi nyata.

Tiga institusi, tiga data, satu gambaran yang sama: "Ancaman terbesar bukan dari AI itu sendiri, melainkan ketika manusia memilih untuk tidak beradaptasi ketika AI terus bergerak maju."