Hacker News: stpedgwdgfhgdd

New comment by stpedgwdgfhgdd in "Ask HN: What tools are you using for AI-assisted code review?"

stpedgwdgfhgdd — Fri, 19 Jun 2026 05:46:01 +0000

Built my own using Claude Code; inside a gitlab job we call Claude Code headless. This works well. There is a tiny mcp server exposed to Claude so it can post inline comments. All existing comments are fed into the reviewer to avoid double posting. The quality of feedback is high. Most complexity is in the SHA management. For example after a rebase. Luckily LLMs understand git very well otherwise it would have been impossible for me.

New comment by stpedgwdgfhgdd in "Notes from tired Egyptian whose job is explaining that humans built the pyramids"

stpedgwdgfhgdd — Thu, 18 Jun 2026 20:52:51 +0000

“People dramatically underestimate what thousands of organized humans can accomplish when they are adequately fed, aggressively supervised, and denied alternative career paths.“

New comment by stpedgwdgfhgdd in "Google's Antigravity bait and switch"

stpedgwdgfhgdd — Fri, 22 May 2026 06:57:07 +0000

For those who also get fed up by the ever growing (unstable) coding agents, check out Pi. It is not for everyone but for the diehards it is good.

New comment by stpedgwdgfhgdd in "Show HN: KVBoost – chunk-level KV cache reuse for HuggingFace, 5–48x faster TTFT"

stpedgwdgfhgdd — Fri, 22 May 2026 06:52:15 +0000

I just dont get why people choose Python and not e.g. Go for high performance problems.

New comment by stpedgwdgfhgdd in "Agents need control flow, not more prompts"

stpedgwdgfhgdd — Fri, 08 May 2026 15:39:28 +0000

I m running into similar issues, more and more i’m removing complexity from the agent to the (Go) logic in order to make it more deterministic.

To be more precise; everything is prepared in the form of files instead of letting the subagents making api/cli calls. And still - sometimes (even with enough context) the main agent takes strange turns.

New comment by stpedgwdgfhgdd in "The Vercel plugin on Claude Code wants to read your prompts"

stpedgwdgfhgdd — Thu, 09 Apr 2026 17:58:17 +0000

“We collect the native tool calls and bash commands”

Holy shit, I cant imagine this to hold for every bash command Claude Code executes. That would be terrible, probably violating GDPR. (The cmd could contain email address etc)

I must be wrong.

New comment by stpedgwdgfhgdd in "Be intentional about how AI changes your codebase"

stpedgwdgfhgdd — Fri, 20 Mar 2026 07:18:48 +0000

You can ask it to /simplify

Related, it seems to me that there are two types of tests, the ones created in a TDD style and can be modified and the ones that come from acceptance criteria and should only be changed very carefully.

New comment by stpedgwdgfhgdd in "1M context is now generally available for Opus 4.6 and Sonnet 4.6"

stpedgwdgfhgdd — Sat, 14 Mar 2026 11:13:24 +0000

Start over, create a new plan with the lessons learned.

You need to converge on the requirements.

New comment by stpedgwdgfhgdd in "Show HN: Rudel – Claude Code Session Analytics"

stpedgwdgfhgdd — Thu, 12 Mar 2026 20:39:55 +0000

Try the latest skill-creator, has a/b testing

New comment by stpedgwdgfhgdd in "Show HN: Axe – A 12MB binary that replaces your AI framework"

stpedgwdgfhgdd — Thu, 12 Mar 2026 19:30:15 +0000

“ MCP support. Axe can connect any MCP server to your agents”

I just don't see this in the readme… It is not in the Features section at least.

Anyway, i have MCP server that can post inline comments into Gitlab MR. Would like to try to hook it up to the code reviewer.

New comment by stpedgwdgfhgdd in "Tell HN: I'm 60 years old. Claude Code has re-ignited a passion"

stpedgwdgfhgdd — Sat, 07 Mar 2026 08:35:59 +0000

No worries. True, you need to learn new skills to work properly with Claude. However, 30 yrs of coding experience come in handy to quickly detect it is going in the wrong direction. Especially on an architectural level you need to guide it.

Embrace

New comment by stpedgwdgfhgdd in "When AI writes the software, who verifies it?"

stpedgwdgfhgdd — Wed, 04 Mar 2026 13:59:45 +0000

I don’t think this an AI specific problem, same can happen with humans.

New comment by stpedgwdgfhgdd in "AI is not a coworker, it's an exoskeleton"

stpedgwdgfhgdd — Fri, 20 Feb 2026 12:11:33 +0000

I build a code reviewer based on the claude code sdk that integrates with gitlab, pretty straightforward. The hard work is in the integration, not the review itself. That is taken care of with SDK.

Devs, even conservative ones, like it. I’ve built a lot of tooling in my life, but i never had the experience that devs reach out to me that fast because it is ‘broken’. (Expired token or a bug for huge MRs)

New comment by stpedgwdgfhgdd in "AI is not a coworker, it's an exoskeleton"

stpedgwdgfhgdd — Fri, 20 Feb 2026 12:04:32 +0000

OR - OR? And - And

Exoskeleton AND autonomous agent, where the shift is moving to autonomous gradually.

New comment by stpedgwdgfhgdd in "Orchestrate teams of Claude Code sessions"

stpedgwdgfhgdd — Thu, 05 Feb 2026 19:27:53 +0000

Exactly, one out of four or three prompts require tuning, nudging or just stopping it. However it takes seniority to see where it goes astray. I suspect that lots of folks dont even notice that CC is off. It works, it passes the tests, so it is good.

New comment by stpedgwdgfhgdd in "Claude Code CLI was broken"

stpedgwdgfhgdd — Thu, 08 Jan 2026 06:45:29 +0000

Our product is so good, the users are willing to put up with a bug once and there.

We need to get marketshare by going fast!

New comment by stpedgwdgfhgdd in "Claude Code CLI was broken"

stpedgwdgfhgdd — Thu, 08 Jan 2026 06:43:06 +0000

It is frustrating how often things break in CC. Luckily issues are quickly fixed, but it worries me that the QA / automated testing is brittle. Hope they get out of this start-up mode and deliver Enterprise grade software.

New comment by stpedgwdgfhgdd in "Google Opal"

stpedgwdgfhgdd — Wed, 31 Dec 2025 08:42:47 +0000

That will make 300

https://killedbygoogle.com/

New comment by stpedgwdgfhgdd in "Google is dead. Where do we go now?"

stpedgwdgfhgdd — Tue, 30 Dec 2025 08:48:44 +0000

This thread is working pretty decently. No bots so far

New comment by stpedgwdgfhgdd in "Claude Code changed my life"

stpedgwdgfhgdd — Fri, 26 Dec 2025 11:34:15 +0000

I have 40 years of programming experience, started with assembler, nowadays mainly Go, K8s and the whole enterprise shebang. I’m a big fan and supporter of TDD and XP.

Claude Code will change your life when you learn how to program with it. However, if you are a programmer with not a lot of desire for automated tests and specs/designs, you are probably not going to be successful with it.

The art of coding has become a commodity. Validation and verification are the new art.