Dear fellow HNers: I said my account has a max karma of 32767, and I'm about to hit that limit. So, for my final post, I'd like to share some of my individual aspirations:

* Be honest. We shouldn't lie, manipulate, dark-pattern, etc. We should invest towards a society that can trust each other, and the first step is our own behavior.

* Be reliable. We should make things that work, follow through on promises, represent prospects in good faith, etc.

* Be helpful. We should help colleagues, random strangers, and society in general. Not maneuver to exploit them.

* Be smart. We're not as smart as we think we are, and we are often dumb. Especially when it comes to topics in which we actually have little experience. Also when we're overconfident about something we "know".

* Be humble. We can make mistakes about even the things we know best, and we should always be learning.

* Be constructive. Others make mistakes, too, and we should help them when they do, as we'd want them to help us or our friends&family, when we make mistakes.

* Be decent. No one person deserves a billion dollars of humanity's wealth. Few people agreed to the rules that enable that: it's not a game that we are winning; it's collusion against the masses, which some people hop onto. Also applies to some smaller sums.

* Be sane. If we find ourselves with way more money than our share, then we say, "oops, how embarrassing", and spread it back around, to everyone. We shouldn't be like the billionaire who instead decides all the poors or immigrants or feminists or trans people or artists or kindergarten teachers or whoever are somehow shortchanging billionaires.

* Be introspective. See sane, above.

That's it for now. See you around. I'll continue to be in the world and in tech industry, but it's time for me to take an HN break. If I come back, it might as well be a new account.

16-year-old found Microsoft bug, got admin access to 17.3T-row databases

Gloriously project runs transformer model written in PDP-11 assembly [video]

Strata: Run a 125B-parameter AI model on a normal gaming PC

Hackers stole US Military records during months-long data breach

There Are Plenty of Laws on the Books to Check the A.I. Giants. Use Them

Host Your First Profitable Meetup or Mini-Conference

Tell HN: Thanks for All the Fish

It may soon be possible to create "mirror life"

Decoupled DiLoCo for Resilient Distributed Pre-Training

Invisible XML (2025)

It's the Kernel's Fault! Custom Page Fault Handling With bpf_fault

CM AI Docking Port: teaching AI agents to knock on the right door

Show HN: Made a digital wall for AI Agents to "tag"

The Router Economy

Show HN: Aartool, which attack chains are complete on your Linux host

USPS to Put Cameras in Trucks That Scan Roads for 'Community Safety'

Agentic GPU Programming for MLSys

Gemini 4 Argon (High): Intelligence, Performance and Price Analysis

Purlin: Separating Orchestration from the Datapath of Collectives

Show HN: Preserve.Chat – Turn WhatsApp chats into sharable links

Gemini Skills Will Replace Gems; Gems Being Discontinued

Anthropic's IPO Prospectus Is a Fucking Doozy

Branch Target Reuse, BTR: New Spectre V2 Attack Targeting JIT Compilers

More screen time is linked to poorer child development, study shows

Vulnerability disclosures double to 10k per month as AI fuels exploitation

What is happening in Kyiv. A drone strike near my University [video]

Aurora PostgreSQL now supports querying of Apache Iceberg and Parquet data

Numerai: Architect an AI scientist to predict the stock market

Show HN: Perspica – A semantic diff for reviewing code

India restores ancient water systems (stepwells)