{
  "version": "https://jsonfeed.org/version/1",
  "title": "Jottings of Vishal - engineering",
  "home_page_url": "https://jottings.vishalvshekkar.com/tags/engineering.html",
  "feed_url": "https://jottings.vishalvshekkar.com/tags/engineering/feed.json",
  "description": "Posts tagged with \"engineering\"",
  "items": [
    {
      "id": "https://jottings.vishalvshekkar.com/jots/5722626097.html",
      "url": "https://jottings.vishalvshekkar.com/jots/5722626097.html",
      "title": "Balancing cache writes-vs-utilization, tool calling back-and-forth, reasoning effort, to improve ...",
      "content_html": "<p>Balancing cache writes-vs-utilization, tool calling back-and-forth, reasoning effort, to improve subjective &amp; perceived sense of performance of an LLM-backed agentic workflow, while having a classifier to route each inference call to a complex hierarchy &amp; classes of LLMs, providers, and wrappers which also reroutes during outages, degraded outputs, and latencies, while ensuring a consistent personality to the end consuming entity is the funnest engineering challenge today.</p>\n",
      "date_published": "2026-09-04T18:33:46.260Z",
      "tags": [
        "llm",
        "agents",
        "ai",
        "engineering"
      ]
    }
  ]
}