Shakahs

This was Google's big announcement at their last I/O conference: That Google would become the agent handling all forms of commerce on behalf of the average person worldwide.

Not familiar with this particular library, but similar libraries use S3 Express One Zone which has write latency <10ms, so you can use that for the WAL and compaction can move data onto other storage classes in the background.

Regular S3 has write latency 100-150ms, which might be fine depending on your workload anyway.

Frontier model providers (Anthropic et al.) gate new accounts to impractically low rate limits and spending caps until clients unlock higher limits with cumulative usage, make equivalent cash deposits up front, or talk to a Sales department to work out some other arrangement. They are handicapped by the postpaid billing model.

It's an administrative burden to sign up very every model provider, and there are many independent inference providers now that serve only open source models.

OpenRouter provides a useful service by allowing easy prepaid model access with much higher rate limits, and they also aggregate different model providers to route queries by price, latency, etc.

The full Claude Code system prompts are extracted every update and posted here, all 670 of them.

https://github.com/Piebald-AI/claude-code-system-prompts/tre...

JSON is okay most of the time, but I loath YAML.

How many Kubernetes administration headaches trace back to the need for automated systems to surgically edit YAML? It’s absurd and YAML may be the worst choice for this use case.

Enterprise adds IAM, logging, and analytics, all of which AWS provides for free or for metered usage without needing an enterprise plan.

If you are paying API rates (not using Max subscriptions) there's no reason to use Anthropic's API directly, the same models are hosted by both AWS and Google with better uptime than Anthropic.

Looks very neat.

You may want to optimize the content serving a bit, since it's currently hotlinking multiple large (30MB) videos at 2K resolution from https://svs.gsfc.nasa.gov.

AWS and GCP both have their own custom inference chips, so a better example for hosting Opus on commodity hardware would be Digital Ocean.

https://react-aria.adobe.com is the new Radix, it provides unstyled components with a heavy focus on accessibility and quality. https://github.com/heroui-inc/heroui is the new Shadcn.

I use it via the Kilo Code extension for VSCode, which is invoking Qwen3-Coder via a Cerebras Code subscription.

https://github.com/Kilo-Org/kilocode https://www.cerebras.ai/blog/introducing-cerebras-code

Sonnet/Claude Code may technically be "smarter", but Qwen3-Coder on Cerebras is often more productive for me because it's just so incredibly fast. Even if it takes more LLM calls to complete a task, those calls are all happening in a fraction of the time.

SlateDB offers different durability levels for writes. By default writes are buffered locally and flushed to S3 when the buffer is full or the client invokes flush().

https://slatedb.io/docs/design/writes/

While your technical analysis is excellent, making judgements about workload suitability based on a Preview release is premature. Preview services have historically had significantly lower performance quotas than GA releases. Lambda for example was limited to 50 concurrent executions during Preview, raised to 100 at GA, and now the default limit is 1,000.

Grok are the first models I am boycotting on purely environmental grounds. They built their datacenter without sufficient local power supply and have been illegally powering it with unpermitted gas turbine generators until that capacity gets built, to the significant detriment of the local population.

https://www.datacenterdynamics.com/en/news/elon-musk-xai-gas...

Reading about mainframes feels very much like reading science fiction. Truly awesome technology that exists on a completely different plane of computing than anything else.

This thinly veiled advertisement claims it's a waste of time to understand the tradeoffs in the models you're using, and you should instead pay them to make those decisions for you. No thank you.

Forcing Twitter to make public posts easily accessible is not at all the same as compelling Apple to hand over your private messages.

Per the ongoing Freedesktop discussion, AWS offered to host but Freedesktop is leaning towards self-hosting on Hetzner so they can control their own destiny and sponsors can contribute cash towards the bill instead of donating hardware.

> https://gitlab.freedesktop.org/freedesktop/freedesktop/-/iss...

Same for me, I use "ar <tab>" in Intellij all the time.

Because they own the CDN and most of the bandwidth is from peering, so it essentially costs them nothing. Netlify on the other hand has to pay per GB to AWS.

Note: This is type stripping only, no transpilation / code generation is done.

Both these issues can be resolved by adding some sample code to context to influence the LLM to do the desired thing.

As the op says, LLMs are going to be biased towards doing the "average" thing based on their training data. There's more old backend code on the internet than new backend code, and Tailwind is pretty dominant for frontend styling these days, so that's where the average lands.

Counterpoint: I also used to fix Lexmark printers as a field tech servicing pharmacies. It was routine to see Lexmark MS711dn printers with page counts in the millions. They did not need more than basic maintenance.

After years of Python and TypeScript, I've started using Java as my default for everything. It's just so much more productive. The ancient developer memes that Java is slow and clunky don't apply anymore.

Postgres also had a long-held reputation for being slow and difficult, but it made incremental improvements for decades and now it's the default choice for databases.

I see Java in the exact same position, as the Postgres of languages.

See also: EDI / EDIFACT / ANSI X12. It was supposed to standardize the way businesses exchange data, but every company implements it differently so integrations take forever, and there's an entire of middlemen offering solutions to make it faster.

Wikileaks released DNC's internal emails, where they made plans to sabotage Bernie.

https://www.nytimes.com/2016/07/23/us/politics/dnc-emails-sa...

Bernie supporters filed a lawsuit against the DNC for disenfranchising them. The DNC argued they operate as a private corporation and are free to pick whomever they want "over cigars in a back room".

> “There’s no right to not have your candidate disadvantaged or have another candidate advantaged. There’s no contractual obligation here . . . it’s not a situation where a promise has been made that is an enforceable promise,” Spiva said.

> The DNC is advancing the argument that any claims to be neutral and fair to all candidates were nothing but “political promises” and are unenforceable by law.

https://www.salon.com/2017/05/13/the-dncs-elephant-in-the-ro...

Counterpoint: Hyundai/Kia has a terrible reputation in the car industry and was probably willing to give Waymo more favorable terms than any other manufacturer. Ford and GM are working on FSD by themselves, so they aren't interested.

Hyundai/Kia have had multiple scandals over the past 10 years due to widespread engine failures, fatal vehicle fires, and becoming a huge theft target due to failing to install even the most basic anti-theft technology (RFID key+immobilizer). Each of these has resulted in recalls and lawsuits.

And besides that, Consumer Reports ranks them #10 on the reliability chart.

https://www.consumerreports.org/cars/car-reliability-owner-s...

Well when you describe it that way, OpenAI also started as a blank text box that gave you a bunch of good answers, and they've already expanded with other services.

I already use ChatGPT as my first go-to stop for certain search queries.

Those issues are absolutely worth discussing, in a reasonable way. Cloudflare isn't the bad actor perpetuating these DDoS attacks, and they aren't forcing website operators to use their services either.