Hacker News

Article URL: https://grigio.org/accuracy-vs-speed-in-local-llms-finding-your-sweet-spot/

Comments URL: https://news.ycombinator.com/item?id=47192855

Points: 1

# Comments: 0

Article URL: https://tobiaskucera.art/en/tichy-brezen/

Comments URL: https://news.ycombinator.com/item?id=47192850

Points: 1

# Comments: 0

Article URL: https://veevid.ai/nano-banana-2

Comments URL: https://news.ycombinator.com/item?id=47192846

Points: 1

# Comments: 0

Article URL: https://www.topgear.com/car-reviews/czinger/21c/first-drive

Comments URL: https://news.ycombinator.com/item?id=47192836

Points: 1

# Comments: 0

EDI used to be a standard method of transferring commercial information between customers. But with the advent of Internet, is EDI still being used? Obviously the EDI infrastructure is now useless with how secure the internet is now. But even just the EDI message format stuffs? Is it being used still?

Comments URL: https://news.ycombinator.com/item?id=47192823

Points: 1

# Comments: 1

Article URL: https://github.com/Rahillasne/Orbit

Comments URL: https://news.ycombinator.com/item?id=47192817

Points: 1

# Comments: 1

Article URL: https://mastodon.social/@netblocks/116147264437940657

Comments URL: https://news.ycombinator.com/item?id=47192808

Points: 4

# Comments: 0

Hi guys, i know this is not topic for this site but i want to hear your voice completly not censored. US and Israel again striking with all assets going down ... This will start new cycle of printing money and once again all of us normal people be sacrificed... Some will lose life some will lose opporunity, but more or less we are all screwed...

Comments URL: https://news.ycombinator.com/item?id=47192803

Points: 3

# Comments: 0

I spent 30+ hours verifying a single brand name. Check domain availability across 6 TLDs, realize the .com is $14K not $12, open TESS for trademark search, check GitHub, npm, PyPI, App Store, Google for competitors — 8+ tabs, every time, for every name. And now I'm shipping faster than ever with AI, so the grind multiplies.

Brandomica runs all of that in one search: - Domain availability with real pricing (Year 1, renewal, 3Y TCO — not the teaser rate registrars show) - Trademark risk (~850K USPTO marks via FTS5, not screen-scraping TESS) - Social handles (GitHub, Twitter/X, TikTok, Instagram, LinkedIn) - Package registries (npm, PyPI, crates.io, RubyGems, NuGet, Homebrew, Docker Hub) - App stores, Google competitor overlap - Linguistic screening (7-language profanity + meaning detection) - Phonetic conflict detection (double-metaphone + Levenshtein against trademark wordmarks) - Safety score (0-100) and Filing Readiness summary It works where you already work: web UI, CLI (npx brandomica check acme), REST API, MCP server for Claude/AI agents. No signup, no paywall. Tech: Next.js 16, Vercel, Turso (SQLite FTS5 for USPTO bulk data), Vercel Domains API for pricing. The trademark index is built from official USPTO bulk XML — ~850K live marks, updated daily via launchd. I built this because verification is the tax on launching, not the product. AI makes generating names trivial. Knowing which one is actually safe to use is the hard part. Try it: https://www.brandomica.com CLI: npx brandomica check yourname Source (MCP server): https://github.com/BRNDMK/brandomica-mcp-server

Comments URL: https://news.ycombinator.com/item?id=47192798

Points: 1

# Comments: 0

Article URL: https://github.com/dj-bolt/django-bolt

Comments URL: https://news.ycombinator.com/item?id=47192769

Points: 1

# Comments: 0

You enter in order, but after a few loops and exits, your neighbors are no longer your neighbors. It’s a tunable Python package for reproducible mixing, with benchmark + determinism tests.

Comments URL: https://news.ycombinator.com/item?id=47192733

Points: 1

# Comments: 0

Article URL: https://redmonk.com/kholterhoff/2026/02/26/generative-ai-policy-landscape-in-open-source/

Comments URL: https://news.ycombinator.com/item?id=47192727

Points: 1

# Comments: 0

If you are still coding the same way you did two years ago, you might be falling behind without realizing it.

Not because you are not capable. Not because you are not using AI. But because you are using AI the same way you did when it first appeared.

A year ago, we talked how much faster AI could make engineering teams. At that time, speed was the headline. Write a requirement, generate code, move faster than before. The improvement was obvious.

After a year of working with AI every day, I’ve realized something more important.

Speed is no longer the advantage. Workflow discipline is.

Today I see two types of engineers.

Both use AI. Both move faster than before.

One uses AI as an upgraded autocomplete. Prompt, review the diff, merge, move to the next task.

The other treats AI as a system component. Before coding, they let the agent clarify the requirement. They separate planning, implementation, and review into explicit phases. They generate tests and extend them with edge cases. They validate critical constraints, the parts of the system where small mistakes create long-term consequences.

The difference may look small in the first few months. Over time, it compounds.

One engineer can only focus on one problem at a time because their workflow is still single-threaded. Prompt, wait, review, merge, repeat.

The other can scale the number of things they build by increasing the number of agents they can clearly define, coordinate, and control.

The difference is not just productivity. It is leverage.

Comments URL: https://news.ycombinator.com/item?id=47192309

Points: 1

# Comments: 1

I've created a comprehensive AI tutorial series covering neural networks, deep learning basics, and practical applications. Perfect for beginners looking to get started with AI development. 16 videos available with more coming soon!

Comments URL: https://news.ycombinator.com/item?id=47192291

Points: 1

# Comments: 0

I was juggling 5+ Claude Code instances simultaneously. Four terminal windows, each with 3-4 tmux panes. "Which Claude is working on which task?" Wasting 10+ minutes just finding the right session.

Built Agent Hand, a Rust rewrite of agent-deck with:

- Ctrl+N jumps to the most urgent session (waiting → ready → running) - Visual status from PTY output patterns (regex, no polling) - Sub-millisecond fuzzy search (Ctrl+G) - Dedicated tmux server (isolated from your default)

The WAITING state (!) is detected from prompt patterns like "Do you want to proceed?" The READY state () uses a TTL—if it was running recently and now idle, it's probably done.

MIT licensed. Would appreciate feedback on the PTY monitoring approach and any edge cases I'm missing.

https://github.com/weykon/agent-hand https://weykon.github.io/agent-hand/

Comments URL: https://news.ycombinator.com/item?id=47192207

Points: 1

# Comments: 0

Article URL: https://github.com/golang/go/issues/77273

Comments URL: https://news.ycombinator.com/item?id=47192203

Points: 2

# Comments: 0

Article URL: https://dontbuild.it/

Comments URL: https://news.ycombinator.com/item?id=47192135

Points: 1

# Comments: 0

Article URL: https://larrycuban.wordpress.com/2026/02/28/students-are-skipping-the-hardest-part-of-growing-up-clay-shirky/

Comments URL: https://news.ycombinator.com/item?id=47192071

Points: 2

# Comments: 0

Article URL: https://prefix.dev/blog/packaging-ai-ml-models-as-conda-packages

Comments URL: https://news.ycombinator.com/item?id=47192061

Points: 1

# Comments: 0

Comments URL: https://news.ycombinator.com/item?id=47192055

Points: 1

# Comments: 0