frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

Ask HN: What Are You Working On? (July 2026)

245•david927•1d ago•956 comments

Ask HN: Should I do a CS masters at Cambridge or start as a new grad at Amazon?

13•sspehr•9h ago•64 comments

Ask HN: Add flag for AI-generated articles

1012•levkk•1d ago•438 comments

Ask HN: Why does Chrome use 100+ mb of memory per tab?

5•Uptrenda•1h ago•1 comments

Ask HN: How has the software industry changed for you personally?

6•brindidrip•55m ago•2 comments

Ask HN: How do you troubleshoot desktop Linux crashes/freezes?

3•Curiositry•4h ago•2 comments

Ask HN: What are your personal websites?

8•basilikum•4h ago•7 comments

Ask HN: AI Agent and harness containerization/security recommendations

2•dv35z•4h ago•0 comments

Ask HN: What is the one YouTube channel you recommend I subscribe to?

3•chirau•1h ago•4 comments

Web Client SDK

2•atoapaymentsapp•6h ago•0 comments

Ask HN: What are the best free ePub books to read?

4•smalltorch•3h ago•2 comments

What's the first thing you look up when evaluating a company?

2•salleisha•6h ago•1 comments

Claudio? Why we need and open source Claude

3•torrmal•7h ago•0 comments

Ask HN: What makes someone good at using Claude Code?

3•Paarthmj•7h ago•7 comments

Finitude of Knowledge

2•Abhishek000001•8h ago•1 comments

Ask HN: How do you see 3D Gaussian Splats could be used in industry?

4•jesseandringa•9h ago•0 comments

Tell HN: Staged NPM publishing is awful

3•electrovir•12h ago•0 comments

Ask HN: How do you use Vim in the era of AI?

48•rstagi•3d ago•95 comments

Ask HN: How do you use LLMs for private discussions?

11•emerongi•1d ago•14 comments

Ask HN: Does appearance matter more than we admit as founders?

5•jmaha•6h ago•6 comments

Let's talk about: LinkedIn ghost jobs

14•xvxvx•1d ago•6 comments

Ask HN: Why isn't Google indexing information about the AT Protocol?

49•iameli•2d ago•33 comments

Anthropic extends Fable 5 access through July 19

4•dboreham•1d ago•2 comments

Agave runs simulateTransaction synchronously on its shared RPC executor pool

2•simonmorley•9h ago•0 comments

What's the best Roblox scrpit Delta or others

2•uussbblbigbois•10h ago•0 comments

How does a Dev's job look like in a few years?

11•korrak•1d ago•20 comments

Omarchy 4 concerns, am I the only one?

4•shivc•12h ago•1 comments

Ask HN: Resources for non-web software architecture?

4•xqb64•1d ago•1 comments

Tell HN: I De-Googled Myself

37•degoogled•3d ago•24 comments

Ask HN: What Did You Work On? (June 2026)

3•adithyaharish•12h ago•3 comments
Open in hackernews

Ask HN: What makes someone good at using Claude Code?

3•Paarthmj•7h ago
I am building Promptster - an AI fluency platform that helps level up engineering organizations. Engineering managers invite their teammates and Promptster analyzes the engineers work with ai coding tools (claude code, codex, cursor, copilot). The manager receives a team-aggregate dashboard where they can roll-out certain practices / skill to their whole team. Each IC receives their own dashboard where they can see their fluency statistics, skill usage, context management, and a DORA dashboard.

We have an open-source rubric that tries to answer, what are best practices for Claude Code (and AI coding agents in-general), would love to hear thoughts and feedback on it. The question is a bit abstract, but I tried my best to figure out practices that I use myself and the top minds agree on.

https://github.com/promptster-ai/rubric

Comments

adamzwasserman•7h ago
1. It is what you do after the prompt: scan diffs in real time and hit the escape key at the slightest sign of trouble.

2. Deep questioning. Constantly probing the assistant: what does it think it is trying to achieve, why did it just make decision X, is there a better way, what does it think the current constraint is?

3. Fighting drift. Knowing that the model will always try to regress to the mean of the training corpus, and constantly being on guard against that drift.

4. Keeping state in your head, because the model cannot. It is up to the programmer to remeber what connects to what else in what way and why.

Paarthmj•7h ago
Thanks a bunch Adam! I think the fighting drift point is super interesting, haven't heard that often!
PaulHoule•7h ago
in general the context goes bad if it gets long enough. if it seems to be losing the thread and you are fighting with it and going in circles: STOP! You might think it would be useful to keep going in one conversation but unless your next task is "do what we just did in this other part of the code" it makes sense to start new conversations for new tasks.
toomuchtodo•4h ago
Are there any artifacts that you could recommend for your fourth point to document the state? An Architecture Decision Record (ADR) or its equivalent perhaps?

https://news.ycombinator.com/item?id=35308838

PaulHoule•7h ago
I can't say that resonates with me, but I'd say this in short.

I often use coding agents when I'm feeling emotionally tired and finding it hard to get into something and could use the support. I think of it as pair programming with a junior who is really smart in some ways but is prone to lapses of bad judgement. Alternately I think of myself as a foreman on a construction site.

I will start out with a paragraph of two about what i think about the situation, what i want to get done, concerns i have, examples where something similar is done in the code that it should look at. I always ask

"Does this make sense? Do you have any questions for me before we proceed?"

and go back and forth until it says it makes sense and it's questions are resolved. Usually i will wind up saying something like

"That's really good, but I have this little problem, what do you think we can do about it?"

If I am not worried about the cost of tokens I am inclined, when debugging, to write up what I understand about the problem to the agent and ask it what it thinks the problem is and then go race it trying to understand the code in the IDE and/or the use the debugger. Sometimes when the problem is simple but in my intellectual blind spot it really wins!

Another thing I think people forget is that agents are good for code understanding so if don't know how to do something or don't understand how it works, just open it in the IDE and start asking questions.

A lot of times people will complain that agents 'aren't listening to them' and I'm going to say often this is really a symptom of "doesn't know how to code", "not good at thinking systematically", "isn't competent as a subject matter expert". My own take is that sycophancy means that sometimes instead of just confronting you and saying "take this job and shove it", models will just act confused. I know often when that is going on I eventually come to realize that I didn't understand the problem and my prompts were garbage so GIGO. When I talked /w Microsoft Copilot about it, it said that LLMs learn to mimic the patterns we use so if you talk like you are confused... it takes like it is confused.

Paarthmj•7h ago
Yeah, the mimicking pattern thing is very interesting, didn't know that but definitely relevant. Love the way you use it as a pair programmer, definitely best if you want to learn! I would love to know why the idea doesn't resonate with you though. Also, thanks a bunch for the detailed feedback!
PaulHoule•4h ago
Well I think one reason I've enjoyed this career for as long as I have is that I want to learn. Like, I had a summer job programming Java applets when Java was still in beta and one thing I love about LLM-assisted programming is that I am still learning new things about Java.

So far as your article goes it seemed poorly organized to me. Like on a site like HN a lot of articles go by and people cannot give every article the attention the author might want, so you have to write things in an "inverted pyramid" style where a few paragraphs at the top lay out the most important points, then you might have a list of principles or heuristics and then you could go through and describe them in details and break out tables about their characteristics. I can't entirely agree with the mindset of Greene's 48 Laws of Power but I think that book is a masterclass in good organization.