knowledge/computer science/ai llm engineering

ai llm engineering

• vllm settings11

• 2026-10-22FlashInfer or Triton for local inference
• 2026-10-22KV cache: FP8 vs NVFP4
• 2026-10-17MXFP4 vs NVFP4
• 2026-10-14Web scraper extraction quality: Trafilatura, readability, Crawl4AI, Firecrawl
• 2026-10-14Web scraper output formats: what each page reader returns for robertz.co
• 2026-10-13Native Claude web search vs self-hosted SearXNG for local LLMs
• 2026-10-13Web page readers for local LLMs: Trafilatura vs Crawl4AI vs Firecrawl vs a Chromium farm
• 2026-10-13Web scraper token cost per page, measured
• 2026-10-13Web search API output formats and token cost for local LLMs
• 2026-10-13Web search quality for local LLMs: did the search engine find the right answer?
• 2026-10-02It isn't SGLang: where the 4× on a DGX Spark actually comes from
• 2026-10-0295 % busy, 1–7 % multiplying: why a sparse MoE leaves GPU compute idle
• 2026-10-01NVFP4 vs 2:4 sparse: where the other half of the petaflop went
• 2026-08-27How to make your non-tech vibe coders team scale, Engineering 3.0
• 2026-06-16RAG vs Fine-Tuning — battle
• 2026-05-30LLM vs VLM vs DIFF — battle
• 2026-05-30Ollama vs vLLM vs Triton — battle
• 2026-05-07INT4 vs NVFP4