Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

A staff engineer's journey with Claude Code

sanity.io

Post reply on HN
Print view View original

191–200 of 410 posts

Re: A staff engineer's journey with Claude Code

#191

Post by searene »

jprokay13 wrote: post #68
lucasyvas wrote: post #55

Earlier quoted context omitted.

I reached this conclusion pretty quickly. With all the hand holding I can write it faster - and it’s not bragging, almost anyone experienced here could do the same. Writing the code is the fast and easy part once you know what you want to do. I use AI as a rubber duck to shorten that cycle, then write it myself.

I am coming back to this. I’ve been using Claude pretty hard at work and for personal projects, but the longer I do it, the more disappointed I become with the quality of output for anything bigger than a script. I do love planning things out and clarifying my thoughts. It’s a turbocharged rubber duck - but it’s not a great engineer

Me too. I’ve been playing with various coding agents such as Cursor, Claude Code, and GitHub Copilot for some time, and I would say that their most useful feature is educating me. For example, they can teach me a library I haven’t used before, or help me debug a production issue. Then I would choose to write the code by myself after I’ve figured everything out with their help. Writing code by myself is definitely faster in most cases.

Re: A staff engineer's journey with Claude Code

#192

Post by adastra22 »

colordrops wrote: post #137
adastra22 wrote: post #99

Earlier quoted context omitted.

My set of Claude agent files have an explicit set of interface definitions. Is that what you’re talking about?

Are Claude agent files per module? If so, then I guess so.

Per source code module?

Re: A staff engineer's journey with Claude Code

#193

Post by nomoreofthat »

EnPissant wrote: post #143
jaggederest wrote: post #101

Earlier quoted context omitted.

I don't know either but I've seen it write to files in plan mode. Very confusing.

That's not possible. You are misremembering.

It’s entirely possible. Claude’s security model for subagents/tasks is incoherent and buggy, far below the standard they set elsewhere in their product, and planning mode can use subagent/tasks for research.

Permission limitations on the root agent have, in many cases, not been propagated to child agents, and they’ve been able to execute different commands. The documentation is incomplete and unclear, and even to the extent that it is clear it has a different syntax with different limitations than are used to configure permissions for the root agent. When you ask Claude itself to generate agent configurations, as is recommended, it will generate permissions that do not exist anywhere in the documentation and may or may not be valid, but there’s no error admitted if an invalid permission is set. If you ask it to explain, it gets confused by their own documentation and tells you it doesn’t know why it did that. I’m not sure if it’s hallucinating or if the agent-generating-agent has access to internal detail details that are not documented anywhere in which the normal agent can’t see.

Anthropic is pretty consistently the best in this space in terms of security and product quality. They seem to actually care about doing software engineering properly. (I’ve personally discovered security bugs in several competing products that are more severe and exploitable than what I’m talking about here.) I have a ton of respect for Anthropic. Unfortunately, when it comes to sub agents in Claude code, they are not living up to standard they have set.

Re: A staff engineer's journey with Claude Code

#194

Post by alexsmirnov »

MangoCoffee wrote: post #26
swframe2 wrote: post #3

Preventing garbage just requires that you take into account the cognitive limits of the agent. For example ... 1) Don't ask for large / complex change. Ask for a plan but ask it to implement the plan in small steps and ask the model to test each step before starting the next. 2) For really complex steps, ask the model to write code to visualize the problem and solution. 3) If the model fails on a given step, ask it t…

I've been vibe coding a couple of personal projects. I've found that test-driven development fits very well with vibe coding, and it's just as you said break up the problem into small, testable chunks, get the AI to write unit tests first, and then implement the actual code

TDD is exactly that I unable to get from AI tools. Probably, because training sets always have both code and tests. I tried multiply models from all major providers, and all failed to create tests without seen the code. One workflow that helps is to create dirty implementation and generate tests for it. Then throw away the first code and use different model for final implementation.

The best way is to create tests yourself, and block any attempts to modify them

Re: A staff engineer's journey with Claude Code

#195

Post by dajonker »

albingroen wrote: post #33
albingroen wrote: post #32

So we’re supposed to start paying $1k-$1,5k on top of already crazy salaries just to maybe get a productivity boost on trivial to semi trivial issues? I know my boss would not be keen on that at least.

And remember. This is on subsadised prices.

Exactly, makes it feel almost like an advertorial for Anthropic, who likely need most customers to pay 1000 bucks a month to break even.

Re: A staff engineer's journey with Claude Code

#196

Post by interstice »

utyop22 wrote: post #88
jprokay13 wrote: post #68

Earlier quoted context omitted.

I am coming back to this. I’ve been using Claude pretty hard at work and for personal projects, but the longer I do it, the more disappointed I become with the quality of output for anything bigger than a script. I do love planning things out and clarifying my thoughts. It’s a turbocharged rubber duck - but it’s not a great engineer

What you're describing is a glorified mirror. Doesn't that sound ridiculous to you?

That's what rubber ducking is

Re: A staff engineer's journey with Claude Code

#197

Post by fragmede »

jaggederest wrote: post #77
rco8786 wrote: post #54

Earlier quoted context omitted.

I feel like I do all of this stuff and still end up with unusable code in most cases, and the cases where I don't I still usually have to hand massage it into something usable. Sometimes it gets it right and it's really cool when it does, but anecdotally for me it doesn't seem to be making me any more efficient.

The key is prompting. Prompt to within an inch of your life. Treat prompts as source code - edit them in files, use @ notation to bring them into the console. Use Claude to generate its own prompts - https://github.com/wshobson/commands/ and https://github.com/wshobson/agents/ are very handy, they include a prompt-engineer persona. I'm at the point now where I have to yell at the AI once in a while, but I touch essen…

How much voice control have you implemented?

Re: A staff engineer's journey with Claude Code

#198

Post by juped »

dingnuts wrote: post #138
ale wrote: post #45

It’s about time these types of articles actually include the types of tasks being “orchestrated” (as the author writes) that aren’t just plain refactoring chores or React boilerplate. Sanity has quite a backlog of long-requested features and the message here is that these agents are supposedly parallelizing a lot of the work. What kind of staff engineer has “80% of their code” written by a “junior developer who doesn…

the kind of engineer who has been Salesified to the point that they write such drivel as "these learnings" instead of "lessons" in an article that allegedly has a technical audience. it's funny because as I have gotten better as a dev I've gone backwards through his progression. when I was less experienced I relied on Google; now, just read the docs

Yeah, the trusty manual becomes #1 at around the same time as one starts actually engineering. You've entered the target audience!

Re: A staff engineer's journey with Claude Code

#199

Post by beefnugs »

astrange wrote: post #155
albingroen wrote: post #32

So we’re supposed to start paying $1k-$1,5k on top of already crazy salaries just to maybe get a productivity boost on trivial to semi trivial issues? I know my boss would not be keen on that at least.

The high salaries make productivity improvements even more important.

If the world wasnt a garbage hole of mis-alignment and planning : The people seeing positivity out of this stuff would be demanding raises immediately, both AI experts and seniors should be demanding the company pay and train juniors as part of their loyalty commitment to the company

Re: A staff engineer's journey with Claude Code

#200

Post by the_hoffa »

namesbc wrote: post #102

Spending $1500 per-month is a crazy wasteful amount of money

That's 18k a year, or about equal or cheaper than "outsourcing", minus the tax and legal ramifications.

I agree it's wasteful, but from a long-form view of what spending looks like (or at least should/used to look like). Those who see 1.5k/month as "saving" money typically only care about next quarter.

As the old adage goes: a thousand dollars saved this month is 100 thousand spent next year.

Post reply on HN