News for hackers. Stuff that matters.

caelinsutch

Re:Orchestrating Agents on Tuesday August 05, 2025 @05:13PM (#44804394)
Some lessons on architecting, organizing, and deploying LLM driven AI agents at scale after 2+ years of engineering work.
Re:The Web Is Broken – Botnet Part 2 on Sunday April 20, 2025 @05:54PM (#43746775)
CAPTCHAS are also quickly becoming irrelevant / not enough. Fingerprint based approaches seem to be the only realistic way forward in the cat / mouse game
If you don't mind leaving VSCode I'm a huge fan of Zed. Doesn't support some languages / stacks yet but their AI features are on-par with VSCode
The easiest solution to this is probably extracting / formatting the content, then running a diff on that. Otherwise you could use snapshot testing algorithms as a diffing method. We use browserbase and olostep which both have strong proxies (first one gives you a playwright instance, second one just screenshot + raw HTML).
how much does this end up costing to run?
Curious how you think about this compared to Electric SQL [1] - I'm currently deciding what sync solution we're going to use for a product rebuild and have been looking at quite a few

[1] https://electric-sql.com/