Harness 2025 vs today vs 2027: from function calls to swarms and async threads
blaizedsouza · x · 2026-11-09
charlespacker sketches the evolution of agent harnesses in three eras: 2025's function calls and system prompts; today's bash loops, model-agnostic design, compaction, context-window management, parallelism and subagents; and a predicted 2027 of pub-sub, "harness-agnostic" setups, message passing, forking, swarms, superagents, async and threads. A concise snapshot of where coding-agent architecture is heading.
More from coding & agent
- dair-ai founder: own your evals — public benchmarks won't tell you what to ship — dair_ai · 2026-11-09
- Open-source video-to-game: agent rebuilds 31 playable games from clips, ~$6 each — songguoxiansen · 2026-11-09
- 'Agent swarms' is funny branding — you wouldn't call coworkers a swarm — round · 2026-11-09
- 179 blocked tool calls were mostly false positives; a re-verification loop is coming — yb2698 · 2026-11-09
- RL-trained harness adapters drop for LFM3, LFM2.5 and TwiL-8B to boost agentic coding — yb2698 · 2026-11-09
- 179 reward-hacking tool calls blocked during agent RL training, just 0.04% of total — yb2698 · 2026-11-09