itz4blitz's profile picture. software developer, systems administrator, cyber ninja.

itz4blitz

@itz4blitz

software developer, systems administrator, cyber ninja.

Tesla makes a cyberquad for kids?! This is awesome.

itz4blitz's tweet image. Tesla makes a cyberquad for kids?! This is awesome.

Why is @cursor_ai making rolling dice sounds whenever committing code today?!


Totally agree—this is exactly what I've encountered and solved in my workflows. This is where vibe engineering really shines, turning chaos into solid process. To dodge weak QA, I've stacked quality gates like: 1) Codex: GPT-5-Codex for planning: It breaks down the work and…

this paper shows that at this point the problems with vibe coding are mostly human hallucinations are rare and barely noticeable in practice the real issue is development driven by instant gratification and weak qa, like skipping tests and relying on llms for verification

alxfazio's tweet image. this paper shows that at this point the problems with vibe coding are mostly human

hallucinations are rare and barely noticeable in practice

the real issue is development driven by instant gratification and weak qa, like skipping tests and relying on llms for verification
alxfazio's tweet image. this paper shows that at this point the problems with vibe coding are mostly human

hallucinations are rare and barely noticeable in practice

the real issue is development driven by instant gratification and weak qa, like skipping tests and relying on llms for verification
alxfazio's tweet image. this paper shows that at this point the problems with vibe coding are mostly human

hallucinations are rare and barely noticeable in practice

the real issue is development driven by instant gratification and weak qa, like skipping tests and relying on llms for verification


Spent some time with @FactoryAI using Droids - not sure what I'm missing here coming from Claude Code, Codex, etc. Can someone share the sauce?


After 24 hours of using @AmpCode, I have to say I'm very impressed. I lost a lot of time in the last week using Claude Code and Codex for some deep integration work. I appreciate how easily AmpCode is able to work through the same problems with ease. More to come!


AI agents have shown me that grep, sed, find, awk, and jq might be the most powerful commands ever built. Old-school text processing remains undefeated.


After a year of relying on Cursor's built in doc indexing, I've discovered better results using @firecrawl_dev to create markdown docs, markdowntoolbox.com to flatten the results, and GPT Pro to create rules files for Cursor / Continue.dev.


itz4blitz أعاد

Reminder that LM Studio is all you need to serve powerful AI to your whole network. Here’s Llama 3.3 70B streamed to my phone. A model that performs like GPT-4 but offline, private, and in your control.


itz4blitz أعاد

I failed with 5+ SaaS businesses in the last 7 years. And you can learn from my mistakes. If you want to build a SaaS business read this: 👇🧵

DsMatie's tweet image. I failed with 5+ SaaS businesses in the last 7 years.

And you can learn from my mistakes.

If you want to build a SaaS business read this: 👇🧵

Loading...

Something went wrong.


Something went wrong.