Hacker News

Article URL: https://alexanderbjoy.com/two-sentence-journal-approaches/

Comments URL: https://news.ycombinator.com/item?id=47053410

Points: 2

# Comments: 0

Article URL: https://atproto.com/blog/new-site-2026

Comments URL: https://news.ycombinator.com/item?id=47053377

Points: 2

# Comments: 1

I’ve been frustrated with the DX of running local agents (OpenClaw) inside Docker. It breaks my VS Code integration and feels clunky.

I spent the last month writing a Kernel-Level Driver (using aya for eBPF on Linux and WFP Callouts on Windows) to enforce security underneath the agent process instead of putting it in a container.

The Architecture:

Network: Intercepts sys_connect to force agent traffic through a local DLP proxy (scrubbing API keys).

FS: Hooks sys_unlink to prevent the agent from deleting files outside ./workspace.

Identity: Binds the process to the TPM so the binary can't be exfiltrated.

The repo is still in early alpha (expect kernel panics), but I wanted to share the WFP/Rust bindings I wrote because they were a nightmare to debug.

Technical Question for HN: Has anyone successfully used eBPF for blocking sys_open calls reliably? I'm hitting a race condition on older kernels."

Why this works:

It admits it causes "Kernel Panics" (Honesty = Trust).

It asks a specific technical question (Invites smart people to comment).

It shares code/bindings (Gives value).

Comments URL: https://news.ycombinator.com/item?id=47053361

Points: 1

# Comments: 0

My friends and I started this project in the summer of 2025 with the initial goal of participating in the ARC Prize Kaggle competition. Early on, we were exploring agentic coding with frontier reasoning models and found that models like o3 and o4-mini could generate high-quality synthetic ARC-style puzzles. Our plan was to use these synthetic puzzles to train a smaller model via agentic reinforcement learning (RLVR with interleaved thinking).

To bootstrap this process, we needed successful solution traces from an open-weight reasoning model for cold-start supervised fine-tuning. That requirement led us to investigate GPT-OSS-120B. While doing so, we noticed something unexpected: simply placing the model into the interleaved thinking regime produced large and consistent score improvements on ARC AGI 2 tasks. We were seeing scores that we didn’t think was possible for a medium sized OSS model.

This observation ultimately shifted the focus of our work as we wanted to find out how universally this observation applies while staying within our resource constraints. We concluded that it applies quite generally, with double digit gains in frontier models too.

Previously, I have read debates about whether ARC AGI 2 is primarily a reasoning benchmark or a visual benchmark. I guess we can now add agentic benchmark to the mix as well!

Detailed writeup: https://pivotools.github.io/pivotools-quarto-blog/posts/agen...

Comments URL: https://news.ycombinator.com/item?id=47053355

Points: 2

# Comments: 0

Article URL: https://github.com/probablyArth/callonce-go

Comments URL: https://news.ycombinator.com/item?id=47053323

Points: 1

# Comments: 0

Hey HN. I built Corral because every time I asked an AI coding agent to "add auth and payments," it hallucinated for an hour and produced broken code. Wrong imports, phantom endpoints, a login page wired to nothing.

The problem isn't the agent. It's that auth-to-billing-to-gating is genuinely hard to wire, and there's no machine-readable spec for how to do it.

Corral is an open-source CLI (MIT) that gives your agent a spec it can read (llms.txt), then scaffolds auth + Stripe billing into your existing project. It detects your framework (Express, Next.js, Hono, Fastify, and 13+ more), embeds into your existing server (doesn't create a new one), and generates working components: profile page, admin dashboard, plan gating, Stripe checkout, usage metering. One YAML config file controls everything.

The agent workflow is 9 commands. Every command supports --json. Errors include a "fix" field. Exit 0 means deploy-ready.

I hardened this across 10 framework/DB combos with blind agent tests: 3 AI models, 3 rounds each, then a 10-agent fleet. Found and fixed real edge cases like Express 4 vs 5 route patterns, Vite proxy ordering, and agents creating duplicate servers instead of embedding into existing ones.

To try it, paste this into any AI coding agent:

Read: https://llama-farm.github.io/corral/llms.txt Add auth and Stripe billing to my app. Built on Better Auth + Stripe. 18 CLI commands, 30+ templates.

GitHub: https://github.com/llama-farm/corral npm: npx create-corral init Docs: https://llama-farm.github.io/corral/

Comments URL: https://news.ycombinator.com/item?id=47053308

Points: 3

# Comments: 4

Article URL: https://guideposts.org/angels-and-miracles/miracles/gods-grace/yes-miracles-really-do-happen-every-day/

Comments URL: https://news.ycombinator.com/item?id=47053302

Points: 1

# Comments: 1

This provider has been almost fully vibe-coded with Claude. I basically told Claude how the data model should look like and let it verify its assumptions on a real YubiHSM device. When the tests initially failed, Claude re-iterated and corrected its assumptions. In total, building it up to this state took me around 16 hours now.

Comments URL: https://news.ycombinator.com/item?id=47053288

Points: 1

# Comments: 0

Article URL: https://www.nifdi.org/videos/siegfried-zig-engelmann-videos/zig-videos-on-instruction-2.html

Comments URL: https://news.ycombinator.com/item?id=47053281

Points: 1

# Comments: 0

Article URL: https://atproto.com/

Comments URL: https://news.ycombinator.com/item?id=47053259

Points: 2

# Comments: 2

I built a simulator that models: • The full write path (WAL, Memtable, Immutable Memtable, L0 flush) • Leveled compaction with cascading merges • Read path with real bloom filter checks (not mocked) • Live write, read, and space amplification metrics

Comments URL: https://news.ycombinator.com/item?id=47053251

Points: 1

# Comments: 0

Article URL: https://www.nytimes.com/2026/02/12/opinion/ai-companies-college-students.html

Comments URL: https://news.ycombinator.com/item?id=47053234

Points: 2

# Comments: 0

Article URL: https://huggingface.co/papers/2602.12670

Comments URL: https://news.ycombinator.com/item?id=47053217

Points: 1

# Comments: 1

Article URL: https://www.complexity-explorables.org/about/

Comments URL: https://news.ycombinator.com/item?id=47053214

Points: 1

# Comments: 0

Article URL: https://github.com/i2y/konpeito

Comments URL: https://news.ycombinator.com/item?id=47052686

Points: 1

# Comments: 0

Article URL: https://newsroom.snap.com/snapchat-launches-creator-subscriptions

Comments URL: https://news.ycombinator.com/item?id=47052675

Points: 1

# Comments: 1

I've been working on creating a Low Power FM radio station for the east San Fernando Valley of Los Angeles. We are not yet on the broadcast band but our channel will be 95.9FM and our range can been seen on the homepage of our site.

KPBJ is a freeform community radio station. Anyone in the area is encouraged to get a timeslot and become a host. We make no curatorial decisions. Its sort of like public access or a college station in that way.

This month we launched our internet stream and on-boarded about 60 shows. They are mostly music but there are a few talk shows. We are restricting all shows to monthly time slots for now but this will change in the near future as everyone gets more familiar with the systems involved.

All shows are pre-recorded until we can raise the money to get a studio.

We have a site secured for our transmitter but we need to fundraise to cover the equipment and build out costs. We will be broadcasting with 100W ERP from a ridgeline in the Verdugos at about 1500ft elevation. The site will need to be off grid so we will need to install a solar system with battery backup. We are planning to sync the station to the transmit site with 802.11ah.

This is a pretty substantial thing involving a bunch of different people and projects. I've built all of our web infrastructure using Haskell, NixOS, Terraform, and HTMX: https://github.com/solomon-b/kpbj.fm

The station is managed by a 501c3 non-profit we created. We are actively seeking fundraising, especially to get our transmit site up and running. If you live in the area or want to contribute in any way then please reach out!

Comments URL: https://news.ycombinator.com/item?id=47052674

Points: 1

# Comments: 0

Article URL: https://www.japantimes.co.jp/environment/2026/01/23/energy/india-cheap-green-tech/

Comments URL: https://news.ycombinator.com/item?id=47052668

Points: 1

# Comments: 0

Scored on 6 metrics — profit, risk-adjusted returns, and consistency. Study what the top earners do differently, then use it to win more yourself.

Comments URL: https://news.ycombinator.com/item?id=47052660

Points: 1

# Comments: 0

Article URL: https://zyroi.com/blog/when-your-ai-swipes-right

Comments URL: https://news.ycombinator.com/item?id=47052655

Points: 2

# Comments: 0