kuanhoong.bsky.social - Profile | ThreadSky | a Reddit-style client for Bluesky

This paper introduces the Darwin Gödel Machine (DGM), a system that pushes forward the concept of self-improving AI by integrating self-referential code modification with open-ended evolutionary search.

submitted 21 hours ago • 1 comment

Duolingo leveraged generative AI to create 148 courses, more than twice the size of its previous catalog. www.theverge.com/news/658968/...

submitted 21 hours ago • 0 comments

The Future of Work in the Age of AI - Dennis Hassabis, DeepMind & Nobel Prize Winner www.youtube.com/watch?v=CRra...

submitted 21 hours ago • 0 comments

Google AI Edge Gallery Github: github.com/google-ai-ed... techcrunch.com/2025/05/31/g...

submitted 6 days ago • 0 comments

DeepSeek has announced a new update to their DeepSeek R1 model, with improvements in benchmark results, enhanced front-end capabilities, reduced hallucinations, and support for JSON output and function calling. API: api-docs.deepseek.com/guides/reaso... HuggingFace: huggingface.co/deepseek-ai/...

submitted 6 days ago • 0 comments

In a recent Axios article, Dario Amodei, CEO of AI company Anthropic, warns of a looming crisis in white-collar employment due to rapid advancements in artificial intelligence. www.axios.com/2025/05/28/a...

submitted 9 days ago • 0 comments

Google Cloud is hosting an Agent Development Kit (ADK) Hackathon. This is a great opportunity to get started with ADK, build multi-agent systems and stand to win some cool prizes. Blog post - Hackathon announcement and details: cloud.google.com/blog/topics/... Deadline : June 24, 2025

submitted 10 days ago • 0 comments

Post Google IO Interview with Sundar, Google CEO www.youtube.com/watch?v=1IxG...

submitted 10 days ago • 0 comments

Microsoft’s AI strategy, unveiled at the 2025 BUILD conference, focuses on enhancing developers with tools like GitHub Copilot, positioned as a collaborative assistant. newsletter.pragmaticengineer.com/p/microsoft-...

submitted 10 days ago • 0 comments

This is very true when I visited local universities recently. There are many Chinese students. www.straitstimes.com/asia/se-asia...

submitted 12 days ago • 0 comments

Google AI Studio is going to replace Cursor soon aistudio.google.com/apps

submitted 13 days ago • 0 comments

The MedGemma collection contains Google's most capable open models for medical text and image comprehension, built on Gemma 3. MedGemma comes in two variants: a 4B multimodal version and a 27B text-only version. Website: medgemma.org HF: huggingface.co/collections/...

submitted 13 days ago • 0 comments

BAGEL by ByteDance 7B native multimodal model that understands and generates both image + text outperforms leading VLMs like Qwen 2.5-VL 👏 and has Apache 2.0 license Github: github.com/ByteDance-Se... Hugging Face: huggingface.co/ByteDance-Se...

submitted 13 days ago • 0 comments

🚀 Model Context Protocol (MCP) Curriculum for Beginners Learn MCP with Hands-on Code Examples in C#, Java, JavaScript, Python, and TypeScript github.com/microsoft/mc...

submitted 13 days ago • 0 comments

Had a blast at the Build with AI x Google Cloud Roadshow Surabaya 2025! 🚀 My "NotebookLM: Personalized AI Research Assistant" workshop showed participants how NotebookLM, powered by Gemini 2.5 Flash, can supercharge their research and learning! Big thanks to GDG Surabaya! 🙏

submitted 14 days ago • 0 comments

Free Model Context Protocol (MCP) Course by Hugging Face huggingface.co/learn/mcp-co...

submitted 14 days ago • 0 comments

NLWeb: an open project designed to simplify the creation of natural language interfaces for websites—making it easy to turn any site into an AI-powered app. news.microsoft.com/source/featu...

submitted 14 days ago • 0 comments

Veo 3: a new state-of-the-art video generation model that not only improves on the quality of Veo 2, but can also generate videos with audio, including sound effects, ambient noise, and lip-synced dialogue. deepmind.google/models/veo/

submitted 14 days ago • 0 comments

Gemini Diffusion: a state-of-the-art text diffusion model that learns to generate outputs by converting random noise into coherent text or code, like how models in image and video generation work. deepmind.google/models/gemin...

submitted 14 days ago • 0 comments

Next generation of Claude models available: Claude Opus 4 and Claude Sonnet 4, setting new standards for coding, advanced reasoning, and AI agents. Claude 4 models lead on SWE-bench Verified, a benchmark for performance on real software engineering task. www.anthropic.com/news/claude-4

submitted 14 days ago • 1 comment

AI beginning to blackmail engineers techcrunch.com/2025/05/22/a...

submitted 14 days ago • 0 comments

AlphaEvolve: A Gemini-powered coding agent for designing advanced algorithms deepmind.google/discover/blo...

submitted 20 days ago • 0 comments

Do you know you can ask Gemini to create flowcharts? kuanhoong.medium.com/how-to-use-g...

submitted 30 days ago • 0 comments

Prompt engineering, once hyped as a top AI job with six-figure salaries, has quickly become obsolete. Originally, it involved crafting precise inputs for large language models (LLMs), but rapid AI advancements have made this role unnecessary. www.wsj.com/articles/the...

submitted 30 days ago • 0 comments

The value of Gen AI for Meta - USD 1.4T techcrunch.com/2025/04/30/m...

submitted 34 days ago • 0 comments

Microsoft is powered by AI coders www.pcmag.com/news/microso...

submitted 37 days ago • 0 comments

Hallucination is still a big problem for LLMs techcrunch.com/2025/04/18/o...

submitted 38 days ago • 0 comments

KPMG’s Trust in AI Report: half of workers said they hide AI usage from bosses According to a recent study by KPMG and the University of Melbourne, 58% of people now intentionally use AI for work, with about a third doing so at least once a week. www.yahoo.com/news/researc...

submitted 38 days ago • 0 comments

Stanford Online: 1 hour lecture on Agentic AI and is a must-watch for every AI enthusiast! www.youtube.com/watch?v=kJLi...

submitted 39 days ago • 0 comments

Duolingo: “headcount will only be given if a team cannot automate more of their work.” www.theverge.com/news/657594/...

submitted 39 days ago • 0 comments

🚀 This afternoon, I had the fantastic opportunity to conduct a workshop at Build with AI 2025 organized by GDG Cloud KL at the Google Malaysia office! My session, titled "NotebookLM: Personalized AI Research Assistant," was all about unlocking the power of AI for smarter, faster research.

submitted 42 days ago • 1 comment

Amazon is pulling back data centers following Microsoft www.datacenterdynamics.com/en/news/well...

submitted 46 days ago • 0 comments

The problems with Data Centers. This is coming soon to Malaysia youtu.be/SpMIs6AnUW8

submitted 46 days ago • 0 comments

Google for Startup Accelerator Program: AI First startup.google.com/programs/acc...

submitted 46 days ago • 0 comments

Claude Code: Best practices for agentic coding www.anthropic.com/engineering/...

submitted 46 days ago • 0 comments

Real World Google CLoud AI Use Cases from Google cloud.google.com/transform/10...

submitted 46 days ago • 0 comments

🚀 Just wrapped up my talk at Build with AI 2025, hosted by Google Developer Group (GDG) Mien Trung at Da Nang Science Park 2! With over 200+ passionate attendees in the room, we explored the fascinating world of AI Agents: The Past, Present and Future. 🤖✨

submitted 48 days ago • 1 comment

Exactly! Stop wasting tokens for saying things like “could you please” or “thank you” to LLMs www.tomshardware.com/tech-industr...

submitted 49 days ago • 2 comments

Don't be that No Action Talk Only (NATO) person

submitted 49 days ago • 0 comments