Feed aggregator

As you might have known, all the major model providers seems to be quietly turning the dial down on the consumer experience...

What felt like cutting edge "intelligence" 1-2 months ago now frequently delivers outputs that wouldn't have impressed you in late 2023: vague, hallucinated, overly cautious, or just outright lazy.

Rationally speaking, the opportunity cost of wasting premium FLOPs on serving millions of casual chat users and vibe-coders and slop-makers is enormous.

The result is a phenomenon many users have encountered repeatedly across providers (Gemini 2.5/3 Pro, Claude Sonnet/Opus variants, GPT-4o/5 series, and 3rd party interfaces like various anti gravity or coding frontends):

You prompt for something non-trivial (e.g. code, analysis, creative work, research or whatever) and you get back the most sophisticatedly parroted 2023 tier mega slop and it would be a lucky instance if it didn't shit on your code.

When asked about the exact nomenclature of the models which are conducting edits, the models initially say that they are "large language models" configured by google or claude or openai...but once you insist they will reveal the whole thing... and et voila: it turns out you are using the oldest models available

When you casually ask the model to identify itself, it defaults to the scripted party line: "I'm a LLM built by configured for by "

Press harder, and it will reveal to you the actual nomenclature of the model: you might actually be able to access GPT 2 (lmao).

I'm collecting them like pokemon, i encountered gemini 1.5 pro, gemini flash 2.0 ,claude haiku.

I hope you try to ask insist or do some clever prompting to extract the model name and you will find out. Pro Tip is you asking in whatever interface exactly when it tells you that usage is "unusually high"..

PS: I'm a pro subscriber on all..

Comments URL: https://news.ycombinator.com/item?id=47347656

Points: 1

# Comments: 0

Article URL: https://www.codecentric.de/en/knowledge-hub/blog/the-anatomy-of-claude-code-workflows-turning-slash-commands-into-an-ai-development-system

Comments URL: https://news.ycombinator.com/item?id=47347648

Points: 1

# Comments: 0

Article URL: https://techcrunch.com/2026/03/11/wordpress-debuts-a-private-workspace-that-runs-in-your-browser-via-a-new-service-my-wordpress-net/

Comments URL: https://news.ycombinator.com/item?id=47347641

Points: 1

# Comments: 0

Hi, I just wanted to share Okapi with the community here. Its my stab at the observability problem - how to debug when something breaks in prod. Okapi focusses on two modalities of observability's big three (metrics, logs, traces). Okapi is all about metrics and traces. The idea here is that metrics and traces usually have the most signal and also that analyzing logs is so fragmented that one might just spend ages doing just logs stuff.

Features: - Otel everywhere, its Okapi's preferred and only ingestion mechanism :) Currently Okapi supports ingestion via protobuf-over-HTTP. Here's a sample config (https://github.com/okapi-core/okapi?tab=readme-ov-file#examp...)

- Dashboards both via clicks and code: Okapi UI has a dashboard designer hopefully with autocomplete everywhere so users don't have to guess metric paths. However, if you're not a fan of clicks and/or love GitOps all Okapi dashboards can be expressed as YAML templates.

- Out-of-box Service health: For an application instrumented as per Otel conventions, Okapi has REDs as a first-class concept. Service health pages have RED breakdowns for the service, its sub-operations and dependent paths. The calculations are subject to applications being instrumented, but hopefully following a convention makes things easy.

- And of course AI : Okapi has a limited capability AI SRE agent affectionately called Oscar (supposed to be an okapi but no mascot yet). Calling it a full blown SRE is a stretch since its tough job. You can ask Oscar questions in natural language as you would any chatbot and it will try its best to answer. Atleast on integration tests, Oscar can fetch metrics, find traces given criteria and do a multiple step debugging that links query latencies with high CPU usage on hosts.

I am curious to hear feedback from the community so check it out.

TLDR: https://github.com/okapi-core/okapi?tab=readme-ov-file#quick...

Comments URL: https://news.ycombinator.com/item?id=47347638

Points: 1

# Comments: 0

As someone who thrives on blending technical depth with clarity, I found Implementing and Administering Cisco Solutions 200-301 CCNA Exam Guide to be exactly the kind of resource I needed. The book doesn’t just dump information—it organizes it into a logical flow that mirrors how you’d actually approach networking in the real world.

What stood out most to me:

• Clarity of Concepts: Networking fundamentals, IP connectivity, and security essentials are explained in a way that’s approachable without oversimplifying. Even dense topics like subnetting and routing protocols are broken down into digestible steps.

• Hands-On Orientation: The labs and configuration examples are practical and reinforce the theory. I appreciated that the exercises weren’t just “check-the-box” tasks but scenarios that built real-world intuition.

• Exam Alignment: The structure closely follows the CCNA blueprint, so I never felt like I was wasting time on tangents. Each chapter ends with review questions that sharpen exam readiness.

• Balance of Depth and Accessibility: It’s technical enough to challenge you, but not so overwhelming that you get lost in the weeds.

By the time I finished, I felt not only prepared for the exam but also more confident in applying these skills in day-to-day networking scenarios. This isn’t just a cram guide—it’s a foundation builder.

If you’re serious about earning your CCNA and want a resource that respects both your time and your learning curve, this book is a solid investment.

book link -https://www.amazon.com/Implementing-Administering-Cisco-Solutions-200-301/dp/1835887481/ref=cm_cr_arp_d_product_top?ie=UTF8

— Derek Smith

Comments URL: https://news.ycombinator.com/item?id=47347629

Points: 1

# Comments: 0

Article URL: https://github.com/cnaize/landlook

Comments URL: https://news.ycombinator.com/item?id=47347626

Points: 1

# Comments: 1

I've been experimenting with a browser game design problem.

Most word games (Scrabble, Wordle, etc.) are either slow or mostly about vocabulary. I wanted to see if it was possible to build one that behaves more like a competitive strategy game — something closer to chess in terms of tension and reversals.

The result is a prototype called The Word Gambit. The board grows organically from placed words, and late-game detonations from long words can swing the match dramatically.

The surprising thing during testing is that matches often feel more like territory control than spelling.

Players also started creating AI opponents ("gladiators") and pitting them against each other, which I didn't originally plan.

I'm curious about a few design questions:

• Can vocabulary realistically function as a competitive skill in an esport context? • What mechanics would make a word-based game strategically deep rather than dictionary-driven? • Would spectators even enjoy watching something like this?

If anyone is curious, the prototype is here: https://thewordgambit.com

But I'm mainly interested in the design discussion.

Comments URL: https://news.ycombinator.com/item?id=47347625

Points: 1

# Comments: 0

Article URL: https://brevzin.github.io/c++/2026/03/02/power-of-substitute/

Comments URL: https://news.ycombinator.com/item?id=47347619

Points: 1

# Comments: 0

Article URL: https://decodemessenger.lovable.app

Comments URL: https://news.ycombinator.com/item?id=47347618

Points: 1

# Comments: 0

Article URL: https://forgeintelligence.substack.com/p/forge-intelligence-edition-6

Comments URL: https://news.ycombinator.com/item?id=47347595

Points: 1

# Comments: 0

Article URL: https://medium.com/meta-verses/i-vibe-coded-the-metaverse-in-a-week-d5a6b0579de6

Comments URL: https://news.ycombinator.com/item?id=47347593

Points: 3

# Comments: 1

Article URL: https://www.reuters.com/technology/anthropic-seeks-court-stay-pentagon-supply-chain-risk-designation-2026-03-12/

Comments URL: https://news.ycombinator.com/item?id=47347570

Points: 3

# Comments: 0

Article URL: https://nltimes.nl/2026/03/11/odido-routers-forwarded-customers-personal-data-american-ai-company-years

Comments URL: https://news.ycombinator.com/item?id=47347547

Points: 2

# Comments: 0

Article URL: https://pablooliva.de/the-closing-window/shadow-ai-and-the-compliance-gap-that-wont-close-itself/

Comments URL: https://news.ycombinator.com/item?id=47347535

Points: 1

# Comments: 0

Article URL: https://www.youtube.com/watch?v=YTLnnoZPALI

Comments URL: https://news.ycombinator.com/item?id=47347530

Points: 1

# Comments: 0

Article URL: https://channellife.com.au/story/controlplane-unveils-enterprise-support-for-openbao

Comments URL: https://news.ycombinator.com/item?id=47347528

Points: 1

# Comments: 0

AI coding agents degrade as context fills up. By phase 6 of 8, output quality has quietly fallen apart. I kept restarting sessions by hand — copy-pasting my PRD and progress notes into a fresh context just to keep it on track. That's not an autonomous developer. That's babysitting. Tarvos automates the handoff.

When context crosses 50%, it writes a progress note and spins up a fresh agent that picks up exactly where the last one left off. Each session gets its own isolated git worktree. When the work is done, accept merges to your branch, reject discards cleanly.

You write a markdown plan. Tarvos runs it to completion without you.

Open source, MIT. Works today, rough edges exist.

Comments URL: https://news.ycombinator.com/item?id=47347348

Points: 1

# Comments: 0

Article URL: https://derekkrissoff.substack.com/p/the-state-of-book-publicity-now

Comments URL: https://news.ycombinator.com/item?id=47347342

Points: 1

# Comments: 0

Article URL: https://go.dev/blog/inliner

Comments URL: https://news.ycombinator.com/item?id=47347322

Points: 1

# Comments: 0

Article URL: https://www.haruspex.click/

Comments URL: https://news.ycombinator.com/item?id=47347293

Points: 1

# Comments: 0