Notes & Insights
Deep dives into machine learning, LLM deployment, agentic AI, and production systems.
Deep dives into machine learning, LLM deployment, agentic AI, and production systems.

A deep-dive into NVIDIA's open-source NemoClaw stack and OpenShell runtime announced at GTC 2026 — how infrastructure-level safety, sandboxed execution, and privacy routing are reshaping how we build and deploy autonomous AI agents.

A hands-on guide to building multi-agent AI systems for enterprise, covering LangGraph orchestration, tool design, sandboxed code execution, and LLM-as-Judge evaluation from real production experience.

A hands-on engineering guide to deploying large language models on-premise using vLLM and llama.cpp, with real-world insights on KV cache optimization, quantization strategies, and throughput tuning from production deployments.