Dev proposes hybrid harness routing cheap classifier models with LLM fallback to cut AI bills
Efficient-Path4172 · reddit · 2026-10-09
A Reddit developer is building a harness tool that routes repetitive classification tasks to ultra-cheap "System 1" classifier models (like Jev, claimed 200x cheaper and 100x faster), triggering a full LLM only when an "unknown" class appears; the LLM's findings are then fed back as new known classes. Proposed use cases include multi-platform social listening (WhatsApp/Facebook/Instagram) for product feedback analytics, smart homes, and robotics. The author plans to open-source the tool if feedback is encouraging.
More from coding & agent
- Dev is vibe coding an open-world RPG for the N64 — Dimillian · 2026-10-09
- Rule of thumb: automate predictable workflows, reserve AI agents for decisions — sujingshen · 2026-10-09
- Shakespeare: an open-source Mac typing-stats app built with Claude Code — whitew0lf · 2026-10-09
- OhMyGame open-sourced: an AI agent builds games and then plays them — sujingshen · 2026-10-09
- Codex turns vague install instructions into real charges once a card is stored — CurieuxExplorer · 2026-10-09
- LLM Wikis: Let an AI Curator Build Growing Knowledge Bases in Your Obsidian Vault — dSebastien · 2026-10-09