Boris Cherny on Prompting: Treat Claude Like a Coworker

X postBoris ChernyOpinion

Boris Cherny expresses surprise at people being astonished by his simple prompting style for Claude. He argues that most tasks do not require rigid scaffolding; users should simply set a goal and speak to the model as they would to a coworker.

Key points

  1. Cherny suggests speaking to Claude naturally, just like talking to a coworker.
  2. Overly prescriptive or heavily scaffolded prompts are unnecessary for most tasks, according to Cherny.
  3. Providing a clear objective is usually sufficient for the model to work out the details.
  4. In a follow-up reply, Cherny noted plans to add logging for cache misses to the /usage command.

Why it matters

As prompt engineering often trends towards complex frameworks, Cherny highlights a minimalist, goal-focused approach that relies on the model's ability to reason out tasks from natural instructions.

Analysis

Cherny advocates treating modern LLMs as collaborative peers rather than systems requiring rigid micro-instructions. His follow-up comment regarding cache misses during extended sessions and updates to /usage also points to ongoing operational improvements in tooling.

Original post

Boris Cherny

@bcherny · 6 October 2026

I am surprised that people are surprised this is how I prompt Claude. Talk to Claude the way you would a coworker. There's no secret to prompting. There's no need to be overly scaffolded or prescriptive for most tasks -- give Claude a goal, and it will figure it out. Back in t.co/J19KzkcOGv

8k likes420 replies5.8k bookmarks603.1k views

View on X
More posts in the thread (1)
  1. @pdbunny369 Interesting. Looking at those long sessions I’m guessing you had some large cache misses. Will add logging so you can see those when you run /usage also.
Selected replies (5)
  • @DeckardBuilds @bcherny It really is just this t.co/iIeDQxUDg9
  • @stillnesshum @bcherny @bot please look at this and write it in Notion for Claude to read
  • @pdbunny369 @bcherny thanks, this was never like this before i could go hours into session without this happening and claude it self would let me know its too long of sessions i had the app crash as well in the past cuz of this issues so i know how to use claude but this was out of my control
  • @RyanSael @bcherny I’ve been telling people my prompts wont be useful to share but people keep asking me. So I started sharing how my prompts look like given how badly worded they were t.co/mJVyoqGM5R
  • @orionintx @filipcodes @bcherny 'talk to it like a coworker' assumes your workplace actually lets that coworker in the room

Mentioned links

Summary generated by AI (Gemini 3.8 Flash, 7 October 2026). It may contain errors – the original source is authoritative.

Related entries

Similar in content, found by the AI search.

  • X post:Anatoli Kopadze

    Advanced Claude Features Beyond Basic Prompting

    Referencing Sam Altman's prediction of ten-person billion-dollar companies, Anatoli Kopadze argues that simply paying for an AI plan and writing basic prompts is insufficient. He shares a comprehensive guide highlighting underutilised Claude capabilities, including Projects, Artifacts, Extended Thinking, Memory, and desktop and browser integrations.

    1.5kbookmarks499.5kviews

    AI & LLMs· Resource

  • X post:MIKE

    Prompting Claude Opus 5: New Rules for Developer Workflows

    MIKE highlights a breakdown by Rahul (@sairahul1) covering Anthropic's official prompting guide for Claude Opus 5, released on 24 July 2026. The guide details how Opus 5 requires a fundamental shift in prompting techniques compared to Opus 4.8.

    1.8kbookmarks159.6kviews

    AI & LLMs· Resource

  • X post:Alamin

    8 Structured Prompts for Software Engineering with Claude

    Alamin (@iam_chonchol) advises against treating Claude like a junior intern with simplistic commands and shares eight role-based prompt templates for software development workflows.

    129bookmarks13.5kviews

    AI & LLMs· Resource

  • X post:Alex Veremeyenko

    Prompt Guide to Stop AI Models Sounding Mannered

    Alex Veremeyenko shares a prompt attributed to Anthropic designed to eliminate overly florid and performative writing in Claude, Fable 5.1, GPT-5.6 Sol, and GPT-6 Astra.

    852bookmarks41.8kviews

    AI & LLMs· Resource

  • Article:Anthropic

    Anthropic Prompting Best Practices for Current Claude Models

    Anthropic outlines prompt engineering best practices for its Claude model line-up, spanning Claude Haiku 4.5 through Claude Fable 5.1 and Claude Mythos 5.1. The guidance covers model-specific nuances, XML structuring, few-shot examples, system roles, and optimal placement for long contexts.

    AI & LLMs· Resource

  • Article:Thariq Shihipar

    The New Rules of Context Engineering for Claude 5 Generation Models

    Anthropic engineer Thariq Shihipar explains how the team trimmed over 80% of Claude Code's system prompt for the Claude 5 generation. Advanced models such as Claude Opus 5 and Claude Fable 5 perform better with less rigid micro-management, benefiting instead from clear interface design, progressive context disclosure, and expressive references.

    AI & LLMs· Opinion

Back to the knowledge base

Let's talk about your project

Get in touch

Locations

  • Mattersburg
    Johann Nepomuk Bergerstraße 7/2/14
    7210 Mattersburg, Austria
  • Vienna
    Ungargasse 64-66/3/404
    1030 Wien, Austria

Parts of this content were created with the assistance of AI.