AI Resource Roundup (24h)

TL;DR

AWS published a cluster of AI infrastructure updates across SageMaker MLflow access. Custom portals, Azerbaijani language model training, plus Resilience Hub and OpenSearch Serverless.
In enterprise AI, Anthropic revenue scale and Glean’s cost-cutting pitch stood out as signals of commercialization pressure.
In Korea, the NAVER D2 event note and memory and semiconductor headlines reflect the broader tech ecosystem mood.

This post is a link archive based on materials collected over the last 24h. It is meant to help you jump into primary sources quickly.

🏛️ Training Azerbaijani language models on Amazon SageMaker AI — AWS Machine Learning Blog
- Why it matters: It is worth reading for a concrete example of handling non-English language model training on SageMaker.
🏛️ Build a custom portal with embedded Amazon SageMaker AI MLflow Apps — AWS Machine Learning Blog
- Why it matters: It is useful because the title alone signals how SageMaker MLflow can be embedded into a custom portal.
🏛️ Streamline external access to Amazon SageMaker MLflow using a REST API proxy — AWS Machine Learning Blog
- Why it matters: It is worth reading for architecture guidance on opening SageMaker MLflow to external access.
🛡️ Introducing the next generation of AWS Resilience Hub for generative AI-based SRE resilience journey — AWS Official Blog
- Why it matters: It matters because it shows AWS’s framing of resilience operations for generative AI-based SRE work.
🛡️ Introducing the next generation of Amazon OpenSearch Serverless for building your agentic AI applications — AWS Official Blog
- Why it matters: It is useful for seeing how AWS positions search infrastructure for agentic AI applications.
🛡️ 모두의 NAVER Engineering Day! D2에서 이어집니다. — NAVER D2
- Why it matters: It helps track the flow of the Korean developer community through a NAVER D2 engineering event update.

🛡️ “메모리가 비싸면 다운받아”…메모리 대란에 20년차 밈 재소환 — 전자신문 AI
- Why it matters: It is worth reading as a signal that memory pricing pressure is spilling into mainstream meme culture.
🛡️ 김정기 신임 공정거래조정원장 취임…“신속 피해구제 강화” — 전자신문 AI
- Why it matters: It is a useful reference for the tone that leadership changes can set around the broader industry environment.
🛡️ 삼성 반도체, 이퇴직률 하이닉스보다 낮아...10배 격차 '통계 착시' — 전자신문 AI
- Why it matters: It is worth reading because the title points to competing interpretations of semiconductor workforce statistics.
🛡️ Anthropic's run-rate revenue hits $47 billion — Simon Willison
- Why it matters: It is worth reading for a fast take on the Anthropic revenue headline through Simon Willison’s lens.
🛡️ Glean’s top line crosses $300M as AI budget-cutting becomes its major selling point — TechCrunch AI
- Why it matters: It helps gauge enterprise demand by showing cost cutting as a core AI product selling point.
🛡️ Claude Opus 4.8: "a modest but tangible improvement" — Simon Willison
- Why it matters: It is useful for a concise quality read on Claude Opus 4.8 when comparing model options.

Checklist for Today:

If your team uses SageMaker MLflow, review the embedded portal and external REST proxy posts together.
If you own agentic AI or SRE operations, check the AWS OpenSearch Serverless and Resilience Hub announcements.
If you manage AI budgets, read the Anthropic and Glean pieces alongside Simon Willison’s model commentary to compare market signals.