Dev builds LLM-free chatbot wired to tools, sparking debate on what LLMs add beyond speed
jasonkneen · x · 2026-09-18
Developer CodingGarden built "jev", a chatbot with no LLM at all: it connects to web search, Wikipedia, weather, Todoist and Home Assistant, decides which tool and args to call based on the prompt, and answers instantly with cited sources and zero hallucinations.
Quoting it, jasonkneen pushes back: the only real difference is speed. LLMs can already hook up to any tools, pick tools from a prompt, and be kept from hallucinating by controlling context — so strip out speed and what's actually different?
More from coding & agent
- Distributed-Slides: an MCP server that compiles agent-written talks into offline presentations — arthurcolle · 2026-09-18
- Vercel skills CLI adds Notion-hosted agent skills, no Git repo required — ivanhzhao · 2026-09-18
- Vercel teams have long written GTM and on-call skills in Notion — ivanhzhao · 2026-09-18
- eve adds automatic model selection: agents pick models by task difficulty before inference — cramforce · 2026-09-18
- DHH on Lex Fridman: Nov 2024 split coding into two universes, agentic coding is transformative — zakkohane · 2026-09-18
- Vercel CLI now deploys static artifacts in under one second, build step skipped — cramforce · 2026-09-18