Hacker News: bicepjai

New comment by bicepjai in "Parallel agents in Zed"

bicepjai — Thu, 23 Apr 2026 14:32:11 +0000

Totally agree, the more parallel agents you spawn the more the probability of you vibe coding rather than guide coding. There comes a point when my mind says just commit and move on, which I fight to not do it.

New comment by bicepjai in "Critical Atlantic current significantly more likely to collapse than thought"

bicepjai — Fri, 17 Apr 2026 16:38:06 +0000

Very true. It’s not about Earth itself, it’s about how it will affect us. Earth has been here all along. She can take anything thrown at her literally, it’s the beings on her that can’t.

New comment by bicepjai in "Opus 4.7 is horrible at writing"

bicepjai — Fri, 17 Apr 2026 15:04:16 +0000

4.7 is unusually verbose

New comment by bicepjai in "Ask HN: How do you maintain flow when vibe coding?"

bicepjai — Fri, 17 Apr 2026 06:41:40 +0000

Same boat. 5 agents is my hard limit, past that I lose the thread and it stops being guided coding, turns into vibe coding. I lean into the context switch instead of fighting it. Plan a large task, kick the agents off, walk away. My rule is whatever I switch to has to be little productive than watching tokens stream or shorts. usually a workout or a research video. Reframes the wait for me. For review I go in other direction: spin up 50 to 100 review agents to audit everything and rank issues by severity. I work through the blockers manually and read the actual diffs. Handles most of the bug compounding problem you mentioned. It’s a kind of flow. Definitely not the same as writing the code ourself, nothing is. Just the closest substitute I’ve found in this era

New comment by bicepjai in "George Orwell Predicted the Rise of "AI Slop" in Nineteen Eighty-Four"

bicepjai — Fri, 17 Apr 2026 06:15:27 +0000

Inspired to come up with these original ideas

New comment by bicepjai in "Claude Opus 4.7 Model Card"

bicepjai — Thu, 16 Apr 2026 14:38:07 +0000

This card is a 272 page report. So now we are redefining names :)

New comment by bicepjai in "Claude Code Routines"

bicepjai — Wed, 15 Apr 2026 03:07:16 +0000

Yes totally agree it’s regurgitating crazy expansive text like book author who needs to publish 10 books a day

New comment by bicepjai in "Claude is getting worse, according to Claude"

bicepjai — Tue, 14 Apr 2026 14:42:39 +0000

I notice recently they push way too many tokens to explain things than they use to. I have not measured it, but working with it every day I can tell.

New comment by bicepjai in "Ask HN: What Are You Working On? (April 2026)"

bicepjai — Mon, 13 Apr 2026 15:52:49 +0000

Working on my own agentic system. It’s been so much fun and learning rust amd svelte along the way. Thanks for the inspiration hacker-news community.

New comment by bicepjai in "Running Gemma 4 locally with LM Studio's new headless CLI and Claude Code"

bicepjai — Wed, 08 Apr 2026 15:42:02 +0000

Totally agree lmstudio headless server on a remote machine but control models from your laptop is an amazing workflow. But Gemma 4 was not a good model atleast in my trials “find me the largest text file in all of the current sub folders” it went on a loopy tool call for ever even with Q8

New comment by bicepjai in "Author of "Careless People" banned from saying anything negative about Meta"

bicepjai — Sat, 04 Apr 2026 16:56:19 +0000

Streisand effect. I didn’t care about the book but I will have listened to this book by this weekend.

New comment by bicepjai in "April 2026 TLDR Setup for Ollama and Gemma 4 26B on a Mac mini"

bicepjai — Sat, 04 Apr 2026 14:40:17 +0000

Yes same experience. Goes into loop mode where is sends same command again and again, till we kill it. This was Q_8 version on lmstudio

New comment by bicepjai in "Claude Code users hitting usage limits 'way faster than expected'"

bicepjai — Fri, 03 Apr 2026 04:17:27 +0000

After spending some time on how Claude code (leaked) tools were written, it makes sense why we constantly hit limits, for all the amazing llm capabilities, CC does not have edit tool and always read before write ( this makes sense) but I expected some surgical precision magic software, seems like other agents like master open code and pi are way better than CC in taken usage

New comment by bicepjai in "Ask HN: What is your dev set up like?"

bicepjai — Fri, 03 Apr 2026 03:09:37 +0000

I tried Claude code with zed and zed will eat memory like crazy (128GB RAM) after long sessions. I gave up on zed after that happening for a month

New comment by bicepjai in "Claude Code runs Git reset –hard origin/main against project repo every 10 mins"

bicepjai — Mon, 30 Mar 2026 16:51:13 +0000

If we are living in an era where software release everyday is the norm, no amount of testing is enough to claim stability. Roll the dice everyday with these beautiful stochastic imitators.

New comment by bicepjai in "I am leaving the AI party after one drink"

bicepjai — Fri, 27 Mar 2026 17:32:02 +0000

All knowledge started as someone's opinion. The goal isn't to avoid opinions, it's to stress-test them. That's exactly what HN is for.

New comment by bicepjai in "Playwright MCP Server with NoVNC Display"

bicepjai — Fri, 27 Mar 2026 16:15:27 +0000

I was tired of being a human button-clicker for UI testing while developing with Claude. On my commute home one evening, I started venting to Claude about the pain: I wanted an AI that could control a browser autonomously while I watched, without hijacking my mouse.

Thirty minutes of validation later, "you are up-to something", "this is a great idea", "this doesn't really exist yet" and "you'd have to build it yourself" and "it's doable but nobody's packaged it up yet"... I finally pushed it to search more aggressively. That's when it found mcp-playwright-novnc, a Docker image that runs Playwright MCP with Xvfb + x11vnc + noVNC, so Claude controls an isolated browser while you watch the whole thing live in your web browser. Your mouse stays completely free. Kudos to the author and Thanks

Playwright MCP Server with NoVNC Display

bicepjai — Fri, 27 Mar 2026 16:15:27 +0000

Article URL: https://github.com/xtr-dev/mcp-playwright-novnc

Comments URL: https://news.ycombinator.com/item?id=47544662

Points: 3

# Comments: 1

New comment by bicepjai in "So where are all the AI apps?"

bicepjai — Wed, 25 Mar 2026 17:03:58 +0000

OpenClaw would disagree :) they are live, in business and by all accounts not built on rigorous engineering, my experience supports it. Sometimes scrappy ships and survives, in the llm era.

New comment by bicepjai in "The math that explains why bell curves are everywhere"

bicepjai — Thu, 19 Mar 2026 03:40:27 +0000

This is one of my favorite philosophical questions to ponder. I always ask it in interviews as a warmup to get their thoughts. I’ve noticed that interviewees often curl up, thinking it’s a technical question, so I’ve been modifying the question one after the other to make it less scary. The interviews are for data scientist roles.