As an AI Architect with experience designing scalable, efficient systems for enterprises and research, I’m always on the lookout for breakthroughs that democratise access to cutting-edge tools. NVIDIA’s latest announcement today has me genuinely excited, it’s a massive step forward in open-source AI. They’re expanding their “open model universe” with new models, datasets, and tools tailored for agentic AI, physical...

I’m opening a new Private LinkedIn group: AI Builders & Developers Network. The goal is simple – a high-signal community for people who actually build, ship, evaluate, or influence AI systems. No hype. No “top 10 tools” spam.  Who I’m inviting for the initial cohort What we’ll do inside the group This group will be monitored and curated to maintain a...

There is something so satisfying about when you pull a model into vLLM using Docker, Inc from Hugging Face and think of the potential opportunities while you watch it come down. Just pulled GLM-4.6V-Flash 0 Z.ai‘s nimble 9B-parameter powerhouse from the GLM-4.6V family! This lightweight, open-source multimodal gem is optimized for local deployment and low-latency apps – no need for massive GPUs like its 106B sibling. Key...

Running Local LLMs is a game changer. Don’t get me wrong – I use all of the mainstream Ai models, switching based on the task. But my real edge? Ollama + Hugging Face – if you know, you know. Most people miss the power here: They are two totally separate libraries, most commercial LLMs are available on Ollama’s library – for...