News for hackers. Stuff that matters.

ChadMoran

Re:GPT-6 Astra on Thursday September 03, 2026 @06:23PM (#49557948)
Sol better than Fable? What? I've found it to basically be on part with Opus and I max out 2 accounts on both providers every week.
Re:Claude Fable 5.1 and Claude Mythos 5.1 on Tuesday September 01, 2026 @05:28PM (#49528491)
My hunch is that much of the model tuning to make it more effective has been for its internal thinking prose. That leaks out into its external writing prose.
Re:Accelerating GPT-5.6 Sol Ultrafast on Thursday August 13, 2026 @06:54PM (#49292770)
The comment you're replying to almost feels like it was written by a bot or am I crazy?
Re:Zed DeltaDB on Wednesday August 05, 2026 @09:41PM (#49191380)
I used to call this developer indulgence.
Re:Kill The Cookie Banner on Sunday July 26, 2026 @12:17PM (#49059606)
What does enforcement of not having these banners look like? Wha tif people just... didn't.
Really sad. I used to love staying connected with people but now I'm fed content from people I don't follow to keep me on the platform. I'm over it.
Re:Migrate from OpenClaw on Thursday June 18, 2026 @07:21PM (#48592923)
You know what, you're right. Somehow between reading the parent comment and this one, I forgot the details of it. My mistake.
Re:Migrate from OpenClaw on Thursday June 18, 2026 @02:47PM (#48589719)
One success case doesn't prove the comment you responded to isn't true.
Re:Migrate from OpenClaw on Thursday June 18, 2026 @02:46PM (#48589712)
OpenClaw feels messy and unpolished. It's powerful but unreliable.

Hermes is focused, tight and does exactly what it set out to do. I already moved to Hermes.

Better doesn't mean perfect.
Advanced agentic prompting.
Re:AI agents that argue with each other to improve decisions on Wednesday April 29, 2026 @11:23PM (#47957704)
100%. I have a /codex skill that shells out to the codex CLI using GPT 5.5 xhigh. My /red-team skill uses that always.
Re:AI agents that argue with each other to improve decisions on Saturday April 25, 2026 @04:37PM (#47904377)
I've been doing this with Claude Code and agent teams.

I have a /red-team skill that will use an agent team to criticize it's own work, grade and rank feedback, incorporate relevant feedback and then start over. It has increased the quality of output.

Re:Is anyone actually using OpenClaw? on Wednesday April 15, 2026 @11:18PM (#47788276)
Monitors my company's support email, searches though code, GitHub etc to try and triage the issue and shoot me a message to give me a head start.

Connects stripe subscription cancellation with PostHog events to see if something was frustrating.

Does the same leading up to when someone subscribes to describe successful paths.

Lots of novel thinking when events happen on its own being proactive.

Just because people aren't talking about it doesn't mean it isn't useful. Think creatively.

Re:Claude Code Routines on Wednesday April 15, 2026 @12:10AM (#47774586)
Agree. I keep my involvement "close to the metal". These higher order solutions seem to cause more noise than provide signal.
Re:An Introduction to Obsidian on Tuesday April 14, 2026 @12:02PM (#47767415)
No but that seems like a read-query/time concern that can be solved rather than your data.
Re:An Introduction to Obsidian on Monday April 13, 2026 @09:45PM (#47760271)
I have an inbox/ folder and a daily/ folder.

Inbox is where all new docs go to, I just use #tags and file references.

Re:An Introduction to Obsidian on Monday April 13, 2026 @07:08PM (#47759154)
I've embraced radical simplicity with Obsidian.

I have an inbox folder that is where all new docs go. Then daily notes. That's it. I tag lines with #thething #theotherthing.

Tagging acts as the organization, lowers the barrier of entry and keeps things discoverable.

Re:An Introduction to Obsidian on Monday April 13, 2026 @07:07PM (#47759145)
This reminds me of people who build elaborate Notion "second brains" that serve no purpose other than to develop/demo them.
Re:Figma's MCP Update Reflects a Larger Industry Shift on Sunday March 29, 2026 @04:16PM (#47566824)
I'd take a cash bet there is soon going to be an enterprise OpenClaw product and every major business is going to demand that their SaaS partners support it.
Re:Figma's MCP Update Reflects a Larger Industry Shift on Sunday March 29, 2026 @12:50PM (#47564826)
SaaS products are headed to where the UI isn't the undifferentiated factor. People have been so busy building a UI that works for everyone so it works for no one. The real value is the data and the workflow they provide. Make the data accessible to agents (MCP, OpenClaw skills, etc0.

People who do that will do well.

Re:Nobody Gets Promoted for Simplicity on Wednesday March 04, 2026 @04:30PM (#47254166)
Hah, well I've done something right. They've let me stay here almost 15 years.
Re:Nobody Gets Promoted for Simplicity on Wednesday March 04, 2026 @04:29PM (#47254165)
Amazon uses a complicated process called "attributed OPS". Meaning you may not be directly responsible for but you contributed in some way.
Re:Nobody Gets Promoted for Simplicity on Tuesday March 03, 2026 @11:50PM (#47243222)
I launched a technical feature on Amazon's retail platform that is responsible for 9 figures worth of revenue. When I launched it, it had no infrastructure. It was a collection of small changes across every core platform (Detail Page, Cart, Checkout, etc).

At first people were like "Well, you didn't do much" but when they saw the value things changed drastically. It's a bit of marketing you have to do to help bring people along.

Often perceived impact is correlated with complexity, sadly.

Re:Why Developers Keep Choosing Claude over Every Other AI on Thursday February 26, 2026 @11:30AM (#47168276)
Model aside, the harness of Claude Code is just a much better experience. Agent teams, liberal use of tasks and small other ergonomics make it a better dev tool for me.
Re:IBM Triples Entry Level Job Openings. Finds Limits to AI on Thursday February 19, 2026 @02:07PM (#47077733)
AI enabled me to ship more features faster, increasing the value to customers. It's that simple.
Re:IBM Triples Entry Level Job Openings. Finds Limits to AI on Sunday February 15, 2026 @11:55AM (#47025227)
Fair. I've had super-linear growth since launching in September. Zero marketing outside of a referral program. People genuinely love what I'm building. I get multiple emails per week about how people appreciate the software and how I send out weekly emails about everything I've launched.
Re:IBM Triples Entry Level Job Openings. Finds Limits to AI on Sunday February 15, 2026 @11:54AM (#47025214)
First of all, thank you. I've always been told I have a back for seeing opportunities others don't.

For me, I always had the ideas and even as a competent engineer, the speed of development annoyed me.

I think folks get annoyed when their reality doesn't match other people's claims. But I have friends who aren't engineers who have launched successful SaaS products. I don't know if it's jealousy or what but people are quite passionate about how it doesn't have productivity gains.

Hell, I remember Intellisense in Visual Studio being a big boon for me. Now I can have tasks asynchronous, even if not faster, it frees up my time.

Re:IBM Triples Entry Level Job Openings. Finds Limits to AI on Sunday February 15, 2026 @12:45AM (#47021359)
This level of toxicity is why I'm reminded to stay away from HN.
Re:IBM Triples Entry Level Job Openings. Finds Limits to AI on Sunday February 15, 2026 @12:37AM (#47021332)
I don't even know what flagged means lol

Older »