Storia in 2 fonti

How OpenAI Optimized AI Agents For Scale

Inside the Harness, API, and Inference efficiency optimizations that support the fast growth of Codex and ChatGPT Work.

Raccontata da

latent.space

newsletter.eng-leadership.com

Confronto fonti

2 prospettive sulla stessa storia

AI · summaries

newsletter.eng-leadership.comStai leggendo3 h fa

How OpenAI Optimized AI Agents For Scale

OpenAI reduced AI agent serving costs by 20% through multi-layer optimizations (speculative decoding, autonomous kernel optimization, intelligent load balancing) supporting 10M+ users. For tech managers, this makes complex agentic workflows (web search, code execution, API orchestration) economically viable in production, fundamentally reshaping infrastructure budget and AI team strategy decisions.

originale

latent.space1 g fa

Codex from 0 to 10M Users: Building ChatGPT Work — Akshay Nathan, OpenAI

Codex hit 10M users in <2 weeks (July 2026); knowledge workers now 20% of base, growing 3x faster than developers. Unified agent harness for coding + knowledge work signals enterprise shift from siloed productivity tools to integrated AI automation—implies budget/team restructuring toward cross-functional workflow agents.

Timeline cronologica

martedì 28 luglio 2026·latent.space
Codex from 0 to 10M Users: Building ChatGPT Work — Akshay Nathan, OpenAI
OpenAI's core product engineering lead on how they are building ChatGPT Work to make AGI accessible to all of humanity: Sites, OpenClaw, Memory, Subagents, Finance, No-Code and…
mercoledì 29 luglio 2026·newsletter.eng-leadership.com
How OpenAI Optimized AI Agents For Scale
Inside the Harness, API, and Inference efficiency optimizations that support the fast growth of Codex and ChatGPT Work.