zurawiki

Bun.sh advertised JSC perf improvements, some listed under the “Various Miscellaneous Optimizations for Real World Use Cases“ section.

How much of an effect those optimizations had the Speedometer score (if any)?

The ~6% + the few ~1% changes don’t add up to the 60% score increase mentioned.

I'm really happy to see we're exploring ways to make SQL more maintainable. TypeScript did a lot for the JS ecosystem, giving us nice autocomplete and making many kinds of bugs impossible to type-check, adding this kind of type-system to SQL makes a bunch of sense.

I selfishly hope this can smooth over all the missing SQL functionality in redshift

Very cool work team!

What about just using Linear or another off-the-shelf task management software?

Seems like most of the features are covered there

Excited to see the progress here on the public launch!

Stacking was a meaningful time-saving as a FAANG sw eng. Excited to see stacking grow within the great tech community

When an organization is large and siloed, there is value in decoupling different parts of the application, assuming the ownership and expertise are properly distributed.

As the comment below mentions, microservices is a fashion–one that took off among overly-staffed eng teams. Having separate parts of the app to deploy can be a liability for a small team frequently deploying breaking changes.

Happy to see the tide turning on microservices. Monoliths make sense for small teams in terms of development and operational complexity.

I was pleasantly surprised to see that AWS offers EC2 pretty beefy instances these days. 64 CPU cores, 128+ GB RAM should scale to most of what a small team would ever need...

In case others find this useful. Just updated for GPT-4, tiktoken-rs is a Rust crate you can use to tokenize text sent to and from OpenAI's LLM models

The idea behind gptee is to bring composability to language models at the shell script level. In line with the unix philosophy, we simplify the LLM interface to just stdin and stdout

https://en.wikipedia.org/wiki/Unix_philosophy

> This idea isn't new, and if you've worked at a company like Facebook, Google, Uber, Dropbox, or Lyft (among others), you've either had first-class support for stacked changes built into your code review platform (i.e. Phabricator at Facebook...

Phabricator got a lot of stuff about code review right (I may be biased from working at FAANG). I'm happy to see someone trying to improve the code review situation on GitHub. When I did use GitHub at the startups I worked at, we were never fully satisfied with the code review experience.

I hope they surface "Phabricator versions" as first-class objects. I never liked it how GitHub just clobbers the history left behind from force-pushes from to branch.

Facebook still delivers ads on their platform sites, and links your browser (not just cookies) to your identity. FWIW, Google isn't that much different