Dragos Roua

In the age of AI, scaling yourself 100x as a programmer is not enough.

iPhone Duo has the potential to become the "book" of the AI age.

What is this AGI, that everybody talks about?

Vibe coders are not coders.

Neolabs like Anthropci and OpenAI sell inference at a certain price per token.

A one-trillion-dollar valuation is no joke.

You can now share your Grok bots.

Grok Bot quick review: it looks simple, but at the same time useful, just like the first iPhone.

Use Grok Build with local and OpenRouter free models alongside Grok-4.6.

In the final episode I talk about open weight model harnesses, or how to actually talk to your local AI.

In this episode I talk about running an open weights model locally.

Episode 2 of the mini series about how to choose an open weights model, this time how to choose an open weights model provider.

How to choose and open weights model: the difference between slow, dumb and laggy - and state of the art inference, like Claude or Grok.

AI watermarking is not baked into model weights, it works at the generation level, by sampling different tokens, paired with a private key.

China is already way ahead in the AI race, more than the Western media make it look.

"AI watermarking now traces Claude text back to you.

We're not in the Singularity yet.

I’m fortunate enough to have witnessed a few key moments in our recent tech history.

See how to combine entry-level AI subscriptions and free models for Fable-level vibe coding, solid limits, and high productivity on an $80 monthly budget.

Publishing ten apps in four months sounds good.

Pre-digital childhood gave you patterns younger developers never learned.

I kept the Claude Code interface I like and routed Grok's model underneath it via proxy.

I shipped 9 iOS apps in 4 months with AI—and learned how to manage AI cognitive overload.

Running local private AI models is now realistic, affordable, and urgent.

Harvesting attention isn't just for big tech.

I stopped prompting my AI agents and started writing loops instead.

After using Claude (Opus 4.6 to 4.8) for roughly 5 months, I'm experiencing with MiniMax 3.

In 2010 I asked if you were living in an illusion.

In 2010 the OS metaphor for self-improvement was an analogy.

I had 9 definitions of success in 2016 and was split between all of them. 10 years later, here's what I actually learned about enough, alignment, and flow.

I wrote 33 ways to overcome frustration in 2010. 15 years later: most of them work, but only for one type.

I wrote about saying no in 2009.

I wrote about the ancestor syndrome in 2016.

In 2015 I argued that life's lack of inherent meaning is liberating.

A 12-year test of The Diamond Cutter's Buddhist business principles.

10 iOS apps in 4 months, built solo with AI.

I wrote 77 reasons to love your life in 2009.

In 2009 I gave 7 reasons to enjoy life.

I wrote a digital nomad primer in 2010.

In 2011 I wrote this list from inside a disaster.

In 2011 I wrote 33 ways to end your day.

I listed 77 things worth doing in 2013 just because I could.

In 2010 I pulled 7 lessons from Kung Fu Panda.

In 2018 the wrong tool was obvious — you could see the mismatch.

The "get and keep" model of motivation breaks down fast.

I wrote a blogging retrospective after 6 months.

Stability and fluidity aren't opposites — they're prerequisites for each other.

In 2018 I bet that technology would outpace ideology as the force reshaping society.

In 2009 I framed social networking versus real life relationships as a match.

I wrote 100 ways to live a better life in 2009, then 100 subposts, then watched it float on the internet for 15 years.