Feed aggregator

Article URL: https://www.wsj.com/articles/how-staying-small-became-ai-startups-biggest-flex-ec127320

Comments URL: https://news.ycombinator.com/item?id=47123160

Points: 1

# Comments: 0

Article URL: https://www.space.com/stargazing/solar-eclipses/why-do-some-places-on-earth-get-far-more-solar-eclipses-than-others

Comments URL: https://news.ycombinator.com/item?id=47123159

Points: 1

# Comments: 0

Article URL: https://www.nature.com/articles/d41586-026-00506-y

Comments URL: https://news.ycombinator.com/item?id=47123155

Points: 1

# Comments: 0

Article URL: https://seths.blog/2026/02/misguided-optimization/

Comments URL: https://news.ycombinator.com/item?id=47123153

Points: 1

# Comments: 0

Article URL: https://www.experimental-history.com/p/the-rise-and-fall-of-peer-review

Comments URL: https://news.ycombinator.com/item?id=47123133

Points: 1

# Comments: 0

After years of building for companies, I finally built something for myself.

Aura Tracker: Habits & Goals is now live on the iOS App Store.

It's an opinionated productivity system that connects Identity → Goals → Habits → Tasks → Deep Work → Insights → Reflection, all in one smooth coherent flow. I built it because most productivity apps I used felt like productivity tools. Also, I was highly influenced by the multitude of books I read on the topic and wanted to build something for myself where I can put all my learnings into use.

PS: This was kinda on my TODO list for like almost a decade. I quit my job last year to go full solo founder mode. As long as I was employed, I used to keep watching others posting Show HN and kept wondering when my time would come. Now that it has, it feels so good. Just the same kind of joy I felt, when I wrote my first "Hello World" program on Turbo C editor as a kid (long ago).

Not sure how this product would do, but the feeling that I built something that's gonna stay forever on the internet (and hopefully change the lives of whoever uses it for the better) is a feeling that no amount of corporate bonus or RSU's can provide !!

Comments URL: https://news.ycombinator.com/item?id=47123132

Points: 1

# Comments: 0

Article URL: https://instadeq.com/blog/posts/elixir-a-low-floor-high-ceiling-language-for-your-projects/

Comments URL: https://news.ycombinator.com/item?id=47123111

Points: 1

# Comments: 0

Article URL: https://instadeq.com/blog/posts/interval-research-corporation-a-1990s-parc-without-a-xerox/

Comments URL: https://news.ycombinator.com/item?id=47123110

Points: 1

# Comments: 0

Article URL: https://www.asimov.press/p/agar

Comments URL: https://news.ycombinator.com/item?id=47123109

Points: 1

# Comments: 0

Article URL: https://asteriskmag.com/issues/13/a-brief-history-of-the-history-of-science

Comments URL: https://news.ycombinator.com/item?id=47123106

Points: 1

# Comments: 0

Hi HN,

When OpenAI started phasing out GPT-4o from the main ChatGPT interface, I noticed a surprising amount of backlash — not about benchmarks or features, but about tone.

A lot of people (myself included) felt that 4o had a certain conversational warmth that later models don’t quite replicate in the same way. That difference was subtle, but noticeable.

So I built a small side project for myself: https://www.ilove4o.com/

It’s a minimal interface that connects directly to GPT-4o via the OpenAI API. No extra layers, no personality hacks — just a focused 4o-only chat experience.

I’m sharing it here because: - There seems to be real user preference around model “personality.” - I’m curious whether others noticed the same behavioral shift. - It raises an interesting question: how much of perceived “friendliness” comes from system prompts, UI, or subtle model tuning?

If you try it, I’d genuinely love feedback — especially from people who’ve spent significant time with multiple model versions.

Happy to answer technical or architectural questions.

Comments URL: https://news.ycombinator.com/item?id=47123105

Points: 1

# Comments: 0

Article URL: https://blog.desigeek.com/post/2026/02/building-microkernel-part0-why-build-an-os

Comments URL: https://news.ycombinator.com/item?id=47123091

Points: 1

# Comments: 0

Article URL: https://www.backblaze.com/blog/announcing-b2-neo-the-storage-problem-neoclouds-dont-talk-about/

Comments URL: https://news.ycombinator.com/item?id=47123071

Points: 1

# Comments: 0

Article URL: https://www.gamesradar.com/hardware/retro/the-gametank-just-smashed-its-crowdfunding-goals-and-im-thrilled-at-the-idea-of-a-new-8-bit-retro-console-in-the-big-26/

Comments URL: https://news.ycombinator.com/item?id=47123068

Points: 1

# Comments: 1

Article URL: https://cagriy.github.io/AI-We-are-asking-the-wrong-question

Comments URL: https://news.ycombinator.com/item?id=47123065

Points: 1

# Comments: 0

Article URL: https://hsph.harvard.edu/news/proximity-to-nuclear-power-plants-associated-with-increased-cancer-mortality/

Comments URL: https://news.ycombinator.com/item?id=47123064

Points: 1

# Comments: 1

I built Fiscal (fscl), a headless CLI for Actual Budget that's optimized for AI agents running in the terminal (Claude Code, OpenClaw, etc).

It acts as an Actual Budget client, so it can sync with an existing Actual server. I built it because I wanted an agent-friendly way to handle repetitive budgeting work while still being able to review everything in the Actual web dashboard.

Site/docs: https://fiscal.sh GitHub: https://github.com/fiscal-sh/fscl

Comments URL: https://news.ycombinator.com/item?id=47123049

Points: 1

# Comments: 0

I needed a CSV library in Zig and I hand rolled one. Then I decided to come back to it and make it avoid allocations entirely and then went down a rabbit hole of performance tuning and learned a ton in the process.

This is the result. I also added a benchmarking library and a blog post that explains the implementation details. All are available in the repo page.

I presented this in a local Zig meetup and it landed well so I figured I'll post it here as well.

Comments URL: https://news.ycombinator.com/item?id=47122443

Points: 1

# Comments: 0

I built Attest because every team I've seen building AI agents ends up writing the same ad-hoc pytest scaffolding — checking if the right tools were called, if cost stayed under budget, if the output made semantic sense. It works until the agent gets complex, then it collapses.

60–70% of what makes an agent correct is fully deterministic: tool call schemas, execution order, cost budgets, content format. Routing all of this through an LLM judge is expensive, slow, and unnecessarily non-deterministic. Attest exhausts deterministic checks first and only escalates when necessary.

The 8 layers: schema validation → cost/perf constraints → trace structure (tool ordering, loop detection) → content validation → semantic similarity via local ONNX embeddings (no API key) → LLM-as-judge → simulation with fault injection → multi-agent trace tree evaluation.

Example:

from attest import agent, expect from attest.trace import TraceBuilder @agent("support-agent") def support_agent(builder: TraceBuilder, user_message: str): builder.add_tool_call(name="lookup_user", args={"query": user_message}, result={...}) builder.add_tool_call(name="reset_password", args={"user_id": "U-123"}, result={...}) builder.set_metadata(total_tokens=150, cost_usd=0.005, latency_ms=1200) return {"message": "Your temporary password is abc123."} def test_support_agent(attest): result = support_agent(user_message="Reset my password") chain = ( expect(result) .cost_under(0.05) .tools_called_in_order(["lookup_user", "reset_password"]) .output_contains("temporary password") .output_similar_to("password has been reset", threshold=0.8) ) attest.evaluate(chain) The .output_similar_to() call runs locally via ONNX Runtime — no embeddings API key required. Layers 1–5 are free or near-free. The LLM judge is only invoked for genuinely subjective quality assessment.

Architecture: single Go binary engine (1.7ms cold start, <2ms for 100-step trace eval) with thin Python and TypeScript SDKs. All evaluation logic lives in the engine — both SDKs produce identical assertion results. 11 adapters covering OpenAI, Anthropic, Gemini, Ollama, LangChain, Google ADK, LlamaIndex, CrewAI, and OpenTelemetry.

v0.4.0 adds continuous evaluation with σ-based drift detection, a plugin system, result history, and CLI scaffolding. The engine and Python SDK are stable across four releases. The TypeScript SDK is newer — API is stable, hasn't been battle-tested at scale yet.

The simulation runtime is the part I'm most curious about feedback on. You can define persona-driven simulated users (friendly, confused, adversarial), inject faults (latency, errors, rate limits), and run your agent against all of them in a single test suite. Is this useful in practice for CI, or is it a solution looking for a problem?

Apache 2.0 licensed. No platform to self-host, no BSL, no infrastructure requirements.

GitHub: https://github.com/attest-framework/attest Examples: https://github.com/attest-framework/attest-examples Website: https://attest-framework.github.io/attest-website/ Install: pip install attest-ai / npm install @attest-ai/core

Comments URL: https://news.ycombinator.com/item?id=47122431

Points: 1

# Comments: 0