Pinned Tweet
저는 대기업 그룹사 전체에 반영되는 적성검사, 인성검사, 면접 개발 전문가입니다. 그러한 대용량 채용 검사의 신뢰성과 타당성 계산을 위한 데이터베이스 설계와 인 데이터베이스 컴퓨팅, 병렬 컴퓨팅을 해왔습니다. 주 스택은 R, pl/pgsql, PostgreSQL, Python, Kubernetes, Luna Modeler 입니다.
4
1,752
배성호 ペ・ソンホ retweeted
Replying to @thsottiaux
"Majority of users"
1. Not available in EU
2. Not available on Plus plans.
Tibo you can do better pls
배성호 ペ・ソンホ retweeted
حيلة سرية تطلق القوة الكاملة المرعبة لـ GPT-6 Astra! 🚨💻💥
هل تعلم أن بيئة Codex تحدد عمل الوكلاء الفرعيين بـ 3 فقط بشكل افتراضي؟ هذا الرقم لا يعبر إطلاقاً عن إمكانيات Astra!
فريق OpenAI اختبر Astra مع 64 وكيل فرعي بالتوازي!
للحصول على هذه القوة الخارقة، أضف هذا السطر فقط لملف config.toml لديك:
[agents]
max_concurrent_threads_per_session = 64
مشاهدة Astra وهو يدير 64 وكيل من Luna يعملون بالتوازي في نفس الوقت هو شيء سينمائي مرعب!
أنجز المهمة خلال 19 دقيقة باستهلاك 5% فقط من حد الباقة! السرعة والكفاءة غيرت قواعد اللعبة بالكامل! 🤯⚡️🔥
배성호 ペ・ソンホ retweeted
Here's how to fix Astra
Astra is a GREAT model, the problem is limits, its THIRSTY
Here's how to fix it:
1. Go to ~/.codex/config.toml
2. Add this:
[features]
context_management = true
Astra remembers the important parts of a long chat so it doesnt reread everything each time
Enjoy ;)
GPT-5.6을 사용해서 IRT 소프트웨어를 만들고 있어요. Anthropic의 Claude는 이상하게 만들더니, 역시 GPT는 논문을 똑바로 읽고 모든 오류를 수정하네요. github.com/ContextualWisdomL…
1
224
배성호 ペ・ソンホ retweeted
Turn any GitHub repository into LLM-ready text!
Simply replace "hub" with "ingest" in a GitHub URL and receive a prompt-friendly text ingest for LLMs.
Gitingest is 100% open-source and provides:
- Directory structure
- A brief summary of the project
- The entire content as LLM-ready text
Plus, it comes with a nice python package and you can run the UI locally!
Stay tuned, I'm working on something really cool with this!✨
Link to the GitHub repo in the next tweet!
_____
Find me → @akshay_pachaar ✔️
For more insights and tutorials on ML and AI Engineering!
배성호 ペ・ソンホ retweeted
이십대 신입시절 선배들 따라 장례식장에 간 적이 있었다. 입 꾹 다물고 어깨너머로 뭘 따라서 하고 나왔는데 내가 방금 뭘 한 건지, 나는 하나도 모르는 것들을 선배들은 어떻게 다 알아서 하는지 어리둥절했다. 그런 예절을 가르쳐주는 책이 있었음 좋겠다는 오랜 생각이 씨앗이 돼서 이제야 책으로
배성호 ペ・ソンホ retweeted
☹️Google Scholar is a great tool. But it doesn't show you how papers are connected with each other.
😀Here's how to fast-track your literature review with visual search using Google Scholar's database:
(And export papers to Zotero, Mendeley, or EndNote.)
배성호 ペ・ソンホ retweeted
LLaVA-o1 is the first visual language model capable of spontaneous, systematic reasoning, similar to GPT-o1!
🤯 11B model outperforms Gemini-1.5-pro,GPT-4o-mini, and Llama-3.2-90B-Vision-Instruct on six multimodal benchmarks.
배성호 ペ・ソンホ retweeted
AWS just released a new Multi-Agent AI framework
It lets you manage multiple AI agents, dynamically route LLM queries, maintain context across AI Agents and can be deployed locally on your computer.
100% opensource.
배성호 ペ・ソンホ retweeted
This is the system prompt for Apple Intelligence. Turns out Apple's prompt engineers are as clueless about how LLM work as all the others.
































