Skip to slide
076 / 170

CH 08 · Parallelism, the Responses API, and the App Server · 01/09

Read as one page

Parallelism, the Responses API, and the App Server

Independent tool calls run in parallel when reads stay concurrent and writes stay serial. Covers the Responses API and App Server.

9 slides · 6 min read

Contour map for Parallelism, the Responses API, and the App Server

In this chapter

  1. 01Parallel tool calls
  2. 02The Responses API: built for agents
  3. 03The App Server: one harness, many surfaces
  4. 04Why it matters
  5. 05How it works: a parallel runner with a dependency guard
  6. 06Code MVP: a parallel tool executor
  7. 07Connecting to the bigger picture
  8. 08Key takeaways

So far the loop has run one tool at a time. But a model often wants several things at once: read three files, run two independent checks. Doing those sequentially wastes time. This chapter covers three connected upgrades: running independent tool calls in parallel, the API shape that was designed for agents (OpenAI's Responses API), and the protocol that lets a single harness power a terminal, an IDE, and a website (the App Server). The first is something you build; the other two are mostly things you understand so you make good choices.

Finished this idea?

Your learning progress stays in this browser.