Ethan Mollick: you're probably not steering your long-running coding agents enough
emollick · x · 2026-09-10
Ethan Mollick argues that a critical factor in successfully using agents in Codex and Claude Code is deciding when and how to intervene. If you aren't steering long-running agentic work, you're likely not managing agents enough — or not building interim reporting that gives you visibility into progress along the way.
Related event: Ethan Mollick on 'driving' long-running agents like Codex(4 posts)→
More from coding & agent
- STM32 Motor Board Running in Under an Hour with Copperkit — debreuil · 2026-09-10
- Cognition launches SWE-2: frontier-level coding performance at up to 70% lower cost — silasalberti · 2026-09-10
- Five key levers for controlling AI costs, and why that advice may soon expire — rseroter · 2026-09-10
- AIRecon: open-source autonomous pentest agent with local Ollama in a Kali sandbox — tom_doerr · 2026-09-10
- Krea Releases Krea Agents, Autonomous Agents Built for Creative Workflows — angrypenguinPNG · 2026-09-10
- KERNEL partners with Stripe's link to let browser agents pay via single-use virtual cards — jeff_weinstein · 2026-09-10