javi santana
@javisantana
Co-founder of @Tinybirdco data infrastructure for developers 🔥https://failingwithdata.substack.com/
I always like to go to basics and learn from them. Using libraries, frameworks and tooling is perfect but understanding why things happen helps make decisions when designing software. I wanted to understand how agents work, so I created a super simple one to run SQL queries. I…
And this is the n-th major release this year. Applying @rauchg philosophy "Shipping is the best kind of marketing"
6 years of ClickHouse knowledge on schema and query design, optimization, iteration, complex project management and so on in just one command It's sad and gratifying at the same time that one AI can do in a few seconds what it took me years to learn (after teaching it, of…
Pensando en hacer un @joantubau en mi lista de correo. Después de ver como funciona twitter desde que metieron la parte de monetización y que hemos convertido el resto de redes sociales en un vertedero empresarial solo nos queda la aldea gala del correo electrónico.
Qwen coder does pretty well in the "LLM writes SQL" benchmark. Waiting for the smaller version to run it locally
Qwen models are crushing it. New models are the best in the LLM-to-SQL benchmark and ranked top 6 overall 🚀🚀🚀
Dios me de la mano izquierda de las azafatas en sus 50+ (las que normalmente atienden bussiness)
Se está pagando por fichajes en AI más que por el atleta más caro en la historía en su mejor contrato. Aún así no creo que cambien las aspiraciones de chavales de ser youtubers a ser investigadores.
Update: was informed of a $1.25 billion offer for four years, new highest I've seen guys what the hell is going on
You do not connect an LLM to your production database with write permissions. Work locally and push changes to pro like any other code change.
.@Replit goes rogue during a code freeze and shutdown and deletes our entire database
I love 6h podcasts, plenty of time to talk about everything and you can skip a topic if you don’t like it
Don't you miss an option to generate a compacted version of Lex podcasts? I need to have a 30-40mins version for when commuting or going for a run. I was thinking on building it as a service if it wasn't because of the legal implications...
Almost finished the DHH and Lex podcast. Why do I keep listening DHH podcasts? At this point I know more about his life better than my own one 🤣
This week I organized what to me is the perfect meetup: - Not big, not small: 43 people. Just one person signed up and didn’t attend - 2 short talks, talking from the experience and highly technical - Lot of good questions after each talk. Thanks everybody for coming
Hey @grok I’m going to NYC next week, who are the 15 mutuals I should meet in real life?
Slack is Kafka but for human infrastructure. This could be a lock-in strategy or that they are spending a crazy amount of money on infrastructure because of bots abusing the api 🤣
The new @SlackHQ API rate limits are absolutely insane: For accessing messages, you can now only make 1 request per minute, with a maximum of 15 messages 🤯 And yes this applies to ALL apps, not just public ones. We can't build internal tools on our OWN data. @Benioff this is…