$ git log --format='%ad %s' --date=short

Archive

23 notes, newest first

2026 23 notes

October

  1. LLM Quantization, Simply: FP16, 8-bit, 4-bit, and What Q4_K_M Means 11 min
  2. Rate Limiting LLM APIs: Count Tokens, Not Requests 12 min
  3. SGLang vs. vLLM: RadixAttention, Structured Outputs, and When to Pick Which 14 min
  4. No GPU? Simulate One: vLLM, llm-d, and 16 Fake H100s on a Laptop 21 min
  5. llm-d, Simply: A Smarter Router for vLLM on Kubernetes 10 min
  6. Temporal Hands-On: Durable Execution on Kubernetes 6 min
  7. Kubernetes GPU Platform Engineering: From nvidia-smi to NCCL 16 min

September

  1. Hermes Agent + LM Studio Bionic: Running Gemma 4 Locally 11 min
  2. TypeSafe’s Jev: Early Access to AI Built for Decisions 11 min
  3. What Is Karmada? Kubernetes Across Clusters, Explained with YAML 15 min
  4. Why LMCache Exists: Stop Making GPUs Read the Same Context Again 11 min
  5. oMLX vs Ollama vs vMLX vs MLX-LM: Local LLMs on Apple Silicon 6 min
  6. AI and LLM Glossary: Common Terms in Plain English 21 min
  7. Why vLLM Is Dominating the AI Inference Market 9 min
  8. What the heck is an AI Agent? 9 min
  9. What Happens When You Ask an LLM a Question 11 min
  10. Run vLLM on Kubernetes with Minikube, WSL2 and NVIDIA GPU 13 min
  11. Running a 35B MoE Model on an 8 GB Laptop GPU: Testing FreeToken 7 min
  12. Kubernetes is Adding a K to YAML: Why KYAML Makes Sense 3 min
  13. Running llama.cpp on a 32 GB MacBook Air: A Direct Comparison with Ollama 9 min
  14. Running Ollama on a 32 GB MacBook Air: A Practical First Setup 7 min
  15. Kubernetes v1.37: An Operator’s Look at Garhwal 8 min
  16. Hello World 2 min