Intel / article

Impressions from visiting OpenAI, Anthropic, & Cursor

Source: newsletter.pragmaticengineer.com
Author: Gergely Orosz
Date: Published Jun 30, 2026

Context

Running a coding remotely lets its work continue independently of a developer's laptop. Gergely Orosz describes this shift in his account of visits to OpenAI, Anthropic and Cursor. The attraction is a separate development environment where an agent can work for longer and the developer can return to inspect results. That changes the platform's responsibilities, too. Orosz reports that Cursor was addressing how agents surface environmental problems and what happens when the machine running a task terminates. Cloud execution removes dependence on one laptop while making progress recovery and remote supervision central engineering problems.

Key takeaways · AI-distilled

At OpenAI, more than 95% of non-engineers use Codex rather than ChatGPT. The coding , not the chatbot, is becoming the general-purpose interface for people who do not write code.
Cloud agents have no way to complain. A local agent surfaces warnings to the human watching it; a long cloud run cannot. Cursor's workaround is to interview the model afterward so it "confesses", then give those confessions to the infra team.
OpenAI bought Ona, formerly Gitpod, because cloud development environments turn out to be the right primitive for agents: secure, persistent sandboxes that outlive one session or one laptop.
Why now: models only became good enough to code unattended around Opus 4.5 and GPT-5.4, windows reached ~1M tokens, and skills matured, and cloud providers finally have spare GPU capacity.
What makes a Slack mention feel like a paradigm shift is not Slack. It removes the per-machine MCP and skills setup and the tool switch, so one typed line starts a run that does not live on your laptop.

Terms in this piece · Glossary

AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.
agent harness — The scaffolding around a model that turns it into a working agent — the loop, the tools it can call, and the rules for when to stop.
context window — The maximum amount of text a model can consider at once — its working memory for the current conversation or task.
MCP — The Model Context Protocol — an open standard that lets any AI assistant plug into any tool or data source without custom integration code.

Key quotes

“At OpenAI, more than 95% of non-engineers use Codex, not ChatGPT.”

“As Codex becomes more capable, its most valuable work is unfolding over hours or days, rather than minutes.”

Read the source newsletter.pragmaticengineer.com