Furkan Kaya

This Chinese developer launched Llama 70B locally on a MacBook on a plane and for a full 11 hours without internet ran client projects. He was sitting by the window on a transatlantic flight with a MacBook Pro M4 with 64 GB of memory. WiFi on board cost $25 for the flight. He declined. No cloud API, no connection to Anthropic or OpenAI servers, no internet at all. Just a local Llama 3.3 70B on bf16 and his own orchestrator script. The model runs through llama.cpp. Generation speed, 71 tokens per second. Context around 60,000 tokens. Memory usage, 48.6 GiB out of 64. Battery at takeoff, 3 hours 21 minutes. And he gave the orchestrator this system prompt before takeoff: "You are an offline orchestrator running on a single MacBook. There is no network. The only resources you have are local files in /Users/dev/work, the Llama 70B inference server at localhost:8080, and a battery budget of 3 hours 21 minutes. Process the queue at /Users/dev/work/queue.jsonl (one client task per line). For each task: draft → run local evals → save artefact to /Users/dev/work/done/. Save context checkpoints every 12 tasks so you can resume after a battery swap. Stop only on empty queue or when battery drops below 5%." So the system knows exactly what resources it is running on. It knows it has no connection to the outside world for the next 11 hours. It knows it has finite memory and a finite battery. It knows the human will not intervene until the plane lands. The system runs in 1 loop. Takes a task from the queue, runs it through inference, saves the artifact, writes a checkpoint. Task after task, just like that. And only when the battery drops below 5% does the orchestrator automatically pause, waits for the laptop to switch to the backup power bank, and continues from the last checkpoint. Here is what the system actually writes in his log during the flight: "saved context checkpoint 8 of 12 (pos_min = 488, pos_max = 50118, size = 62.813 MiB)" "restored context checkpoint (pos_min = 488, pos_max = 50118)" "prompt processing progress: n_tokens = 50 / 60 818" "task 37016 done | tps = 71 s tokens text → /Users/dev/work/done/proposal_westside.md" Outside the window, clouds, blue sky, and no WiFi. On the tray, 1 MacBook, an open terminal on 2 screens, and an inference server on localhost. From what I have observed, this is the cleanest offline AI workflow I have seen in the past year: 11 hours of flight, $0 for WiFi, and the entire client queue closed before landing.

Opus 4.7 토큰 테스트 토크나이저 차이로 제미나이의 2배를 사용합니다. Opus 4.6 대비해서도 50% 많이 사용해요. 이건 사실상 같은 한도에서 모델이 50% 더 비싸진겁니다.

CLAUDE CODE'A BAŞLAMAK için izlemeniz gereken videolardan birine TÜRKÇE ALTYAZI ekledim!!! Eğer sağda solda dönen Claude Code laflarını görüp de, ya ben nerden başlayacağım diye düşünüyorsanız, bu video işinize yarayacak. En temel 12 tane özelliği, olabilecek en yalın hali ile anlatmış olan bu video gerçekten kafanızda bir çok konunun oturmasını sağlayacaktır. İlk başlarda ben de, nasıl başlayım, ne izleyim diye çok zaman kaybetmiştim. Ve karşıma çıkan bu videonun başlangıç aşamasında biri için çok faydalı olacağını düşünüp altyazı ekledim. Ha öyle oto caption da değil, baya emek verildi... Bu tarz videolar ile başlangıç aşamasını sizin için kolaylaştırabileceğim bir videolar serisi oluşturmayı planlıyorum ilgi olursa. İyi seyirler. İzleyin, kaydedin, başlayın... 🙋‍♂️

CLAUDE OPUS 4.6 THINKING REDUCED BY 67% - Data shows Claude Opus 4.6 now thinks 67% less than before, dubbed “AI shrinkflation” - Same price but noticeably dumber; users report more guardrails and restricted output - Anthropic stayed silent until public data dropped; suspected compute-saving for next model (Mythos)

This monetization strategy is genius Habitica makes $20K/mo turning your to-do list into an RPG Here's how they do it 👇 Onboarding - Create your pixel character (instant ownership) - Pokemon vibes: a guide walks you through everything - Asks about YOU, not your goals - No paywall. No trial. Fully free to use. Monetization (hidden genius) - Paywall is buried in settings. You'd never find it. - Two revenue streams: gems (€0.99-€22.99) + subscription - Gems = costumes, quests, pets (cosmetic dopamine) - Subscription sells MORE rewards, not access - "Subscribing supports our small team" (guilt + love) - 4 tiers, yearly pre-selected with "Popular" tag + strikethrough price They don't sell the app They sell the loot $20K/mo from people who pay to dress up their to-do list avatar Watch the full flow

Gemma 4 can run on phones without an internet connection! 🤯 It can perform local agentic tasks, such as logging and analyzing trends. When connected, it can also make API calls. Want to try it yourself? Get the Google AI Edge App on iOS or Android. (🔊 Sound on for the demo!)

Bruh, instead of building one app making $19.1k/month, build 191 apps making $100/month like this guy. 🫨

Ücretsiz eğitim vereyim sizlere. B2b pazarlama öğrenmek isteyip de nereden başlayacağını bilmeyenlerin en büyük hatası parça parça bilgi toplamak. Bir gün seo bir gün abm bir gün funnel derken kafada çorba oluşuyor. Oysa iş geliştirme satış altyapısı ve ihracat pazarlama tek bir sistemin parçaları. Bunları ayrı ayrı öğrenirsen bağlantıyı kuramazsın. 35 sayfalık bir B2B eğitim seti hazırladım. Sıfırdan altyapı kurmak isteyenler için. Account based marketing funnel tasarımı organic growth stratejileri hepsini birbirine bağlayarak anlatıyorum. Türkiyede bu konuda düzgün kaynak bulmak zaten işkence herkes genel genel konuşuyor kimse sistemi baştan sona göstermiyor. Ben gösteriyorum abiler. Postun altındaki linkten ulaşabilirsiniz. Umarım işinize yarar🫡

Google’s Gemma 4 E2B running on-device on iPhone 17 Pro Gemma 4 is built from the same research as Gemini 3, has image understanding capabilities and can reason if needed Running at ~40tk/s with MLX optimized for Apple Silicon

To create a great Ad... The problem needs to resonate with the audience. If it doesn’t, they will keep scrolling. Starting with a direct statement or question can create this instant connection.