#agentic

Spec-Driven Development, explained by cats. #cat #code #ai #spec #agentic

AI agents are getting wallets now, but who’s checking if they’re trustworthy? #ai #tech #agentic #claudeai

My multi agent Claude code system experiments on itself in a sandbox overnight #aiagents #ai #claudecode #claude #agentic

90% dos projetos de IA falham. 🤖❌ Não por falta de tecnologia. Por falta de fundamento. Existe um padrão em quase todo projeto que chega até mim com problema: 👉🏿 Timeout sem saber a causa 👉🏿 Agente dando resposta errada com confiança 👉🏿 Base de conhecimento desatualizada e ninguém percebeu 👉🏿 Retrabalho no meio do projeto do zero O motivo é sempre o mesmo: ninguém pensou no Dia 0. Arquitetura no Dia 0 🏗️ Padrões no Dia 0 📐 Observabilidade no Dia 0 👁️ Avaliação no Dia 0 ✅ Sem isso, o projeto cresce em cima de uma base que não aguenta. Quando as alucinações aparecem, e vão aparecer, você não tem como diagnosticar. Aí o caminho é um só: jogar tudo fora e começar de novo. Vibe coding funciona no experimento. 🧪 Na produção, o que sustenta é fundamento. #IA #EngenhariadeDados #AIAgents #DataEngineering #OWSHQ #AIDaDataEngineer #Agentic #MLOps

Agentic AI Roadmap . . . #Agentic #AIEngineer #AI

Agentic Şirket Ne Demek? Yapay zekânın sahaya inmesiyle birlikte teknoloji ve iş dünyası liderlerinin ağzından düşürmediği bir öngörü var: Artık agentic şirketler dönemine girdik. Ya da agentic olmayan şirketlerin yaşama şansı yok. Maalesef, bu kavramın tam Türkçe karşılığı olmadığı için İngilizcesini olduğu gibi kullanıyoruz. Agentic şirketler, yapay zekâ ve otonom sistemleri entegre ederek kendi kendine karar alabilen ve proaktif eylemler gerçekleştirebilen organizasyonlara verilen genel ad. Geleneksel işletmelerden farklı olarak, bu şirketler süreçlerini optimize etmek ve verimliliği artırmak için agentic yapay zekâ (AI) teknolojilerini kullanıyor. Dünya Gazetesi'ndeki köşemde yazdım, okuyunuz ve siz de görüşlerinizi, deneyimlerinizi bildiriniz, lütfen…👇🏻 https://www.dunya.com/kose-yazisi/agentic-sirket-ne-demek/765788 Ulaşılabilir linkler ayrıca ufuktarhan.com ana sayfada ve Instagram profil linkimde, story’de. #futurist #ufuktarhan #agenticcompany #agentic #ai @dunyagazetesi @dunyagazetesii http://dunya.com #DünyaGazetesi #dünya

Comment PAL below if you want to try it out ‼️ This is weird… but I didn’t expect AI to feel like this. I tried PALs by Tavus and it wasn’t the features that caught me off guard… It was the moment it remembered something I said days ago, and checked in without me asking. Not reminders I set. Not commands I gave. Just… context. Feels less like using software and more like talking to something that actually gets you. #TavusPartner #MeetThePALs #ai #ad #agentic

3 courses to learn Agentic AI- Beginner to Advanced . . . {Agentic ai, ai agents, certification, upskill, ai} #agentic #agenticai #trending #ai #tech

Claude just launched Cowork, and it marks a clear shift toward agentic AI. This isn’t chat-based AI anymore. With Cowork, Claude can access real folders on your system, files, screenshots, notes, receipts and operate inside real workflows. . . . [claude.Anthropic.AI.Agentic.Automation] #claude #ai #agentic #automation #tech

There is an accelerating gap between what AI agents can do and whether you can trust them to do it. AI agents are getting more capable fast. They’re also failing in ways that don’t show up on benchmarks. METR, a Berkeley nonprofit that evaluates frontier AI, has been tracking the length of task agents can finish autonomously. That metric has been growing exponentially, with capabilities doubling roughly every 4.3 months. Meanwhile, Arvind Narayanan at Princeton recently released “Towards a Science of AI Agent Reliability,” which decomposed reliability into 12 dimensions borrowed from nuclear and aviation safety. His core finding is that nearly two years of rapid capability progress have produced only modest reliability gains. Capability is on an exponential curve. Reliability is on a flat line. The gap between them is accelerating. That’s the problem anyone deploying agents in real work needs to understand. #ai #aiagents #agentic #aitrust #stayhuman

What if you never had to ask your AI assistant the same question twice? Futurist Ian Khan explains exactly how that works. #Heat #agentic #AI #claude #chatgpt

and we won in the google agentic ai hackathon! 🙌❤️ #hackathons #agentic #firebase #bangalore #womenintech