george_ciobanu

If you had to build a context window manager in 24h, would you stick to the existing model or come up with something better? Here's what I did: 1. Built a proxy that intercepts Codex's calls to OpenAI and rewrites them…

Hi HN, I'm George Ciobanu (https://www.linkedin.com/in/georgeciobanunyc). I built pandō ('CAD for code') because I got tired of watching AI agents burn tokens, take forever, and still get it wrong. Here's (one reason)…

You may have read this story on HN or somewhere else: fisherman hangs out at the beach all day, and the rich man works hard all his life to get to the same place…

what is this: schema = { "actor" => { "direct_fields" => [{ "actor_id" => "smallint unsigned" }, { "first_name" => "varchar(45)" }, { "last_name" => "varchar(45)" }, { "last_update" => "timestamp" }], "primary_key" =>…

https://human.software <- don't know how to make this a link, my apologies. Only supported DB as a data source is Postgres. Note 1: to create account enter your email and press Login/Signup. It will create a new account…

I often find myself having to wait up to multiple hours (in one case days) waiting for an ML model to train or to tune some hyper parameters. I recently discovered Dask and with small changes to my Python code I saved…