Pablo Sanderson

always building.

currently @RevolutApp, prev @SpecterHQ and @eigen_tech. founder @foodtobento

[Pablo Sanderson @psiesta11]

@tibo_maker https://t.co/fPsPXTVERD

[Pablo Sanderson @psiesta11]

@hieudinh_ 👋

[Pablo Sanderson @psiesta11]

📜

[The Dor Brothers @thedorbrothers]

We just made a $200,000,000 AI movie in just one day.

Yes, this is 100% AI.

[Pablo Sanderson @psiesta11]

@Trulli__UB @bajoput El representante

[Pablo Sanderson @psiesta11]

@T3_RM @dvassallo Gran Canaria

[Pablo Sanderson @psiesta11]

@demirdjiantwins NANO

[Pablo Sanderson @psiesta11]

@ainarsklavins @levelsio @Hetzner_Online @digitalocean @TermiusHQ ssh back in and 'claude --continue'

[Pablo Sanderson @psiesta11]

@antinertia arcards

[Bryan Johnson @bryan_johnson]

Longevity pill saved the life of an old (80-90 yr human equivalent), ill German Shepherd. Human trials next.

The pill stabilizes telomeres, the caps holding tips of our DNA from getting loose.

[Pablo Sanderson @psiesta11]

@tdinh_me Good burden to have 🫣- try both with small scoped experiments? 1-2 days feature and then evaluate over time?

[Pablo Sanderson @psiesta11]

@carlos10_ga @OpinionUd @canarias7 @UDLP_Oficial Estaría bien que le preguntaran en alguna de las ruedas de prensa que quedan. Por qué no rota después de los malos resultados?

[Pablo Sanderson @psiesta11]

@dvassallo @levelsio Talk to the mirror style for practising and ask for feedback.

[Pablo Sanderson @psiesta11]

@marvr_ @dobroslav_dev I think so, 2024 at least

[Pablo Sanderson @psiesta11]

@nico_jeannen This so much. Time to put some guardrails before deployment though 🤌🪇 + basic logging

[Pablo Sanderson @psiesta11]

🌴

[Andrej Karpathy @karpathy]

New (2h13m 😅) lecture: "Let's build the GPT Tokenizer"

Tokenizers are a completely separate stage of the LLM pipeline: they have their own training set, training algorithm (Byte Pair Encoding), and after training implement two functions: encode() from strings to tokens, and decode() back from tokens to strings. In this lecture we build from scratch the Tokenizer used in the GPT series from OpenAI.

[David Bressler 📊 @bresslertweets]

Few tips:

- only run in markets where you get most of your subscribers
- A/B test landing pages
- constantly work towards getting your quality score to 10 to reduce CPCs
- start with exact and phrase match searches only
- run search on the Google search only, don’t open up to other channels and search partners

DM me if you need any help

[Pablo Sanderson @psiesta11]

@levelsio the map is missing canary islands - the spot with most hours of sunshine in Europe ☀️☀️☀️

[@levelsio @levelsio]

"You're just lucky"

[Pablo Sanderson @psiesta11]

@javilopen brutal javi! Roughly how many iterations (or time?) did it take for the GPT4 side of things?