---
date: "2024-07-29T20:16:18.000Z"
title: "2024-07-29"
tags: ["protobuf"]
draft: false
---

import { Tweet } from 'astro-embed';

<Tweet
  theme="dark"
  id="https://twitter.com/jxnlco/status/1817593682662408610"
/>

This point resonates with me.
The more time I spend prompting models, the more it's becoming clear that the clarity of the instructions are what matter most.
Writing clear, unambiguous instructions is not easy.
Decrease scope and you have a chance of doing it well.

---

[Protobuf](https://protobuf.dev/) messages/schema in prompt instructions seem to be able to provide similar output structure to Pydantic, as [`instructor`](https://github.com/jxnl/instructor) and other libraries do.
I've experimented with this approach [in the past](/posts/2023/protobuf-contracts-with-llms) but there is a bit more that could be done to structure the messages so that they could serve the dual-purpose of prompt instructions and as RPC schemas.