Boris Cherny
Creator of Claude Code at Anthropic.
Boris Cherny did not write this page.
We collected these quotes from things they published elsewhere, and every quote links to where it was said. They have no account here and have not endorsed this site. Quotes are word for word; the short line under each one is our own restatement, not their wording. Their own site. Is this you? Claim it or ask us to remove it. Or tell us what is wrong here.
Filter All topics · Newest first
-
Our reading
When there is new data, you have to update your priors, even when the resulting churn feels thrashy.
Their wordsWhen there’s new data, you have to update your priors.
-
Their wordsI love being wrong. It is my favorite, because it helps me more clearly define the problem, find the right solution, learn more quickly, and solve the problem.
- 8 weeks earlier
-
Their wordsthe model does not seem to be prompt injectable anymore.
↗Boris Cherny: We Cut 80% of Claude Code's Promptyoutube.com 1st of 20 in this recording
-
prompt injectionneural networks
Their wordsit's it's literally we're looking at neurons in the model's brain that light up when prompt injection happens. So the model won't even tell you but we can actually see those neurons and we can figure out and diagnose that it's happening. And then you combine that with the auto mode classifier and with these three layers we just cannot demonstrate prompt injection anymore.
↗Boris Cherny: We Cut 80% of Claude Code's Promptyoutube.com 2nd of 20 in this recording
-
Their wordsAnd the reason is every model is very different. So, something that you did for one model maybe 3 months ago, it just might not translate at all to the next model.
↗Boris Cherny: We Cut 80% of Claude Code's Promptyoutube.com 3rd of 20 in this recording
-
Their wordsAnd what's interesting is that the model is actually a little bit more intelligent without these prompts.
↗Boris Cherny: We Cut 80% of Claude Code's Promptyoutube.com 4th of 20 in this recording
-
coding agentsAnthropicsimplicity
Their wordsfor people that aren't building agentic products, but you're using Claude code, every 6 months delete your Claude MD. Delete your skills. Delete your hooks.
↗Boris Cherny: We Cut 80% of Claude Code's Promptyoutube.com 5th of 20 in this recording
-
Their wordsAnd you don't want to guess what's the instruction that the model needs because you might not predict it correctly. The thing that you want to do is you want to run it.
↗Boris Cherny: We Cut 80% of Claude Code's Promptyoutube.com 6th of 20 in this recording
-
Their wordsI I think this is sort of the crazy thing about building on models. It's just so different than all the engineering that I've ever done. Like in the past when you built on systems, you built these like big beautiful systems and you really think about the system design up front. You have like a big suite of unit tests. You think about everything and you know, like a re-architecture is a big project.
↗Boris Cherny: We Cut 80% of Claude Code's Promptyoutube.com 7th of 20 in this recording
-
Their wordsAnd the model is not like that. It's um the way to think about it is almost like a like a living creature, like it's something more organic.
↗Boris Cherny: We Cut 80% of Claude Code's Promptyoutube.com 8th of 20 in this recording
-
Their wordsI think evals, they outlive the harness a little bit, but not by that much. Like an eval might live for maybe one, two, three model generations, but nowadays the you know, we're on the exponential. The model is improving so quickly, very often we just saturate the eval, and then we have to throw it away, and we have to come up with a new eval.
↗Boris Cherny: We Cut 80% of Claude Code's Promptyoutube.com 9th of 20 in this recording
-
Their wordsBecause the model can do this at every given model generation, but there is often not a product that lets the model do this and lets it express this kind of ability to do this.
↗Boris Cherny: We Cut 80% of Claude Code's Promptyoutube.com 10th of 20 in this recording
-
Their wordsOne is you should give the model slightly harder tasks than what you think it can do.
↗Boris Cherny: We Cut 80% of Claude Code's Promptyoutube.com 11th of 20 in this recording
-
Their wordsone example is the model can now rewrite essentially any code base from one language to a different language.
↗Boris Cherny: We Cut 80% of Claude Code's Promptyoutube.com 12th of 20 in this recording
-
Their wordsAnd my hypothesis is there's probably dozens, hundreds of opportunities like this with the models of today that no one has yet realized.
↗Boris Cherny: We Cut 80% of Claude Code's Promptyoutube.com 13th of 20 in this recording
-
Their wordsI think the skill nowadays is less about prompt engineering and more about figuring out how do you give Claude a hard task that seems a little bit too hard. And then how do you make it possible for Claude to verify its work along the way? And the verification I think is probably the single most important thing that people do not get right
↗Boris Cherny: We Cut 80% of Claude Code's Promptyoutube.com 14th of 20 in this recording
-
Their wordsYou don't need slash goal, you don't need slash loop. These help, but really all you need is give the model the task, give it a way to verify the output of its work so it doesn't get stuck, and it will just go.
↗Boris Cherny: We Cut 80% of Claude Code's Promptyoutube.com 15th of 20 in this recording
-
Their wordsSo when I look at engineers that have been, you know, coding for a long for a long time, you know, like for for years or for decades, this is a really really common failure mode is trying to over specify and it's trying to be overly specific and then, you know, get the model to do the to do the task exactly the way that you would have done it. And that that's just not the way the model works.
↗Boris Cherny: We Cut 80% of Claude Code's Promptyoutube.com 16th of 20 in this recording
-
Their wordsLike this is actually like a new form of test time compute. Like when we talk about the scaling laws and kind of we talk about the model getting more intelligent over time, historically, it's been a function of the size of the neural net, the amount of training data, and the number of flops that you put in to the training. And then recently, we also added test time compute. So this is essentially a fancy way a researcher way of saying how many tokens does it generate. And now dynamic workflows are essentially a new way to orchestrate test time compute.
↗Boris Cherny: We Cut 80% of Claude Code's Promptyoutube.com 17th of 20 in this recording
-
Their wordswe're on the path to fully automating the maintenance of our apps by doing this. And this is again hundreds of agents running every day, sometimes thousands of agents every day. It's doing the work of you know, dozens or hundreds of engineers.
↗Boris Cherny: We Cut 80% of Claude Code's Promptyoutube.com 18th of 20 in this recording
-
Their wordsSo coding is solved for the kind of coding that I do. It's not solved for everyone. You know, there's still code bases that are like super deep systems code bases where quad still struggles.
↗Boris Cherny: We Cut 80% of Claude Code's Promptyoutube.com 19th of 20 in this recording
-
Their wordsAnd it's really about being empirical. So forget all of the things that you learned about past models. Forget everything that you've learned about computer science theory in class. Look at the model, try to do a task, see where it struggles, and then based on that adjust. So it's just like very much become it's not a theoretical science, it's become an empirical science.
↗Boris Cherny: We Cut 80% of Claude Code's Promptyoutube.com 20th of 20 in this recording
- 5 months earlier
-
Our reading
The scarce skill is now context switching across parallel agents, not sustained deep work.
Their wordsIt’s not so much about deep work, it’s about how good I am at context switching and jumping across multiple different contexts very quickly.
↗Building Claude Code with Boris Cherny — The Pragmatic Engineernewsletter.pragmaticengineer.com
-
Their wordsonce there is a good plan, it will one-shot the implementation almost every time.
↗Building Claude Code with Boris Cherny — The Pragmatic Engineernewsletter.pragmaticengineer.com
-
Their wordsThere’s just no way we could have shipped this if we started with static mocks and Figma or if we started with a PRD.
↗Building Claude Code with Boris Cherny — The Pragmatic Engineernewsletter.pragmaticengineer.com
-
Their wordsalways make sure that when you start a migration, you finish the migration.
↗Building Claude Code with Boris Cherny — The Pragmatic Engineernewsletter.pragmaticengineer.com
- 7 days earlier
-
Their wordsAt Anthropic, we don't build for the model of today, we build for the model of six months from now. And that's still my advice to founders that are building on LLMs.
↗Y Combinator, quoting Boris Cherny on the Lightcone Podcastx.com
- 20 months earlier
-
open sourceJavaScriptdeveloper tools
Their wordsNPM and NodeJS should do more to make ES Modules easy to use
↗NPM and NodeJS should do more to make ES Modules easy to useborischerny.com 1st of 3 in this piece
-
JavaScriptsoftware performance
Their wordsCommonJS can be challenging to statically analyze, and uses an inefficient, synchronous module loading algorithm at runtime.
↗NPM and NodeJS should do more to make ES Modules easy to useborischerny.com 2nd of 3 in this piece
-
Their wordsWe should make type=module the default for new package.json files for the npm init, yarn init, and pnpm init commands.
↗NPM and NodeJS should do more to make ES Modules easy to useborischerny.com 3rd of 3 in this piece