d0100

An employee just got phished by adding a number to a legitimate deviceAdd login route that bypasses 2FA and adds a device with full access to office and mail

Probably working as intended...

My workflow for production code:

1. Pick the most complete project boilerplate (fullstack JS can easily introduce security bugs, SPA + API is best as cheap linting solves most problems)

2. Project skills (how to CRUD without mess)

3. Use worktrees for concurrent features, local session for conflicts

4. Local session for QA and refinement

I use Copilot and GPT 5.4

Managed to shorten pre-AI priced ongoing projects to 2 weeks or a month

GPT 5.4 instead of 5.5.

I am exclusively using 5.4 because its only 1x and very good, but the github calculation showed my once $40 become a $680 billing

That is too expensive and not worth paying

Python is, thus, clearly superior to other languages in terms of ease of review.

Do we get visual comparisons along with this bold claim?

I remember doing a MailChimp "clone" in Laravel some 12 years ago and implementing an email builder using their templates

I am using TS sandboxed in deno for all our agent code generated from a UI builder (inspired by OpenAI's own agent builder, and spits out the same code output)

Are you using Temporal with distributed workers?

We have a simple worker setup and temporal is pretty easy to setup

Out only issue is really needing an intermediary data store for task result storage

We are using DBOS in new projects as it's even simpler and the downside (task log interface behind saas) is easily remedied with a copilot generated task viewer

Tried using Zed but for some reason the AI can't open the browser?

The trick is to constrain the LLM to program in a very defined coding style

If I make the LLM generate code that follows my own code architecture and style, that should be enough creative input

I went from plan with opus, implement with claude, to simply plan and implement with GPT 5.4

It's a very good model for a very good price

GPT 5.4 is working pretty well for me, both in Copilot and Codex vscode extensions

If you create a plan it follows it closely

Any example more complex in the backend?

Are we supposed to expose all entities and relationships and rely on row level security?

AI assisted development is the generalists dream

Althought it still hasn't solved procrastinating the next plan prompt

In a certain way, that's how Brazilian culture works

That's how Brazilian media conglomerate culture works

I still prefer Deno because I don't need a separate npm install command, which makes running code that much easier

Being able to import from "https://my-vpn.com/mypackage@1.2.3" or "npm:package@1.2.3" and just running code without having to worry about scaffolding node_modules makes sandboxing code much easier

In pt-br it's "leito", "semi-leito", "executivo" and "convencional"

"Leito" sounds more premium than "cama", despite being both valid words for "bed"

Why is there no RTX 5060ti?

I wonder if joining layers from the "organs" of different models could further enhance the results

I've shaved off 30s of queries by transforming json columns into a string after the first CTE is done with it

If you apply Pareto to productivity, you can fire a lot of people and still manage

I haven't had much success using LLM's in Unity as there is a lot of work that goes into the editor and cannot be done via code, especially if you are using several different assets

How many crabs where drawn?

Does this UI work with Open Code?

In simple projects where database is only changed via an API, we just audit the API instead. It's easier to display and easier to store than tracking each DB change a single transaction does

My gf manages to get paid using Cursor/Copilot, despite not being able to branch herself out of a loop

In my experience Copilots work expertly at CRUD'ing inside a well structured project, and for MVPs in languages you aren't an expert on (Rust, C/C++ in my case)

The biggest demerit is that agents are increasingly trying to be "smart" and using powershell search/replace or writing scripts to skimp on tokens, with results that make me unreasonably angry

I tried adding i18n to an old react project, and copilot used all my credits + 10 USD because it kept shitting everything up with its maddening, idiotic use if search replace

If it had simply ingested each file and modified them only once, it would have been cheaper

As you can tell, I am still salty about it

I just had a deja vu, I'm sure I read this some months ago

Did you previously write this exact comment before?

Its also small enough that a simple "Claude, migrate to Rust" would work to succesfully launder all the code

Manus is pretty "big" in the entrepreneur crowd here in Brazil

When it came out is was very good, and had much better results than ChatGPT

dbos.dev should take care of long-running tasks, but I haven't tried running too many tasks yet as my other bigger projects all use Temporal

The answer is always money, just ask all the small, rich European countries that have no need for draconian measures against their citizens