Long Context Tests: Infinite Scroll Works Well in Personal Agents
vivekhaldar · x · 2026-08-29
Vivek Haldar questions whether concerns about 'lost in the middle' and context rot are still valid. Based on his personal experience using infinite scroll within agents, it seems to work fine, though he notes this is anecdotal. He suggests that a key reason for success is the main thread's ability to spawn sub-agents to handle heavy lifting.
Related event: Long Context Works Well in Practice as Sub-Agents Share the Load(2 posts)→
More from coding & agent
- LangChain founder: only lab-independent harnesses can survive across models — BraceSproul · 2026-08-29
- Tencent Hunyuan Hy4 Preview Integrates OpenCode Go with 1M Context — TencentHunyuan · 2026-08-29
- Spider-Man game built in 9 Hy4 iterations over 6 hours for just $13, open sourced — TencentHunyuan · 2026-08-29
- Grok Bot Demo: Automating Local Deal Finding and Seller Negotiation — brandon_galang · 2026-08-29
- Cursor to drop GPT model support starting Nov 12 — zainhas · 2026-08-29
- Geek Builds AI Trading Firm Using a Team of 10 Grok Bots — adamamcbride · 2026-08-29