Thiago Preihs

Entrepreneur. Technology and finance.

[Thiago Preihs @thpreihs]

@elonmusk Thanks for everything man

[Thiago Preihs @thpreihs]

@roderix1966 @opapoeconomico Hahahahaha cade o Beto?

[D @dshan]

This is just

so so good.

[Thiago Preihs @thpreihs]

@dshan 👏👏👏

[Thiago Preihs @thpreihs]

"Nirvana was for angry teenagers, Alice in Chains was for angry dads" - by someone on X

[Thiago Preihs @thpreihs]

I just discovered my philosophical DNA. My strongest tendency: Hegelian. Take the diagnostic: https://t.co/8317wI5FS4

[Arpit Bhayani @arpit_bhayani]

Google just released TimesFM (a Time Series Foundation Model) - a 200M-parameter model that can forecast time-series data it has never seen before, with no additional fine-tuning required.

Time-series forecasting is required everywhere - retail, finance, healthcare, etc. And for the longest time, this was the domain of traditional statistical methods. Then deep learning models came along and did better, but they involved long training and validation cycles before you could even test them on new data.

TimesFM changes this. All we need to do is point it at a new dataset, and it gives you a solid forecast immediately - zero-shot.

The architecture is decoder-only, the same idea as GPT. Instead of words, it works with "patches" - groups of contiguous time-points treated as tokens. The model predicts the next patch from all the ones before it.

The model was pre-trained on 100 billion real-world time-points, mostly from Google Trends and Wikipedia Pageviews - which naturally capture a huge variety of patterns across domains.

On benchmarks, zero-shot TimesFM matches PatchTST and DeepAR that were explicitly trained on those datasets, and even beats GPT-3.5 on forecasting tasks despite being far smaller.

The model is open on HuggingFace and GitHub if you want to try it.

[Thiago Preihs @thpreihs]

@roderix1966 @FreeClecio Based 😎💪

[Thiago Preihs @thpreihs]

@Motion_Viz Workflow please :)

[Thiago Preihs @thpreihs]

@roderix1966 @Ricardo_Cd_Oliv Na lata!

[Marc Andreessen 🇺🇸 @pmarca]

The multi decade dream of a high scale prediction market that helps our imperfect selves discern the future has finally arrived.

[Thiago Preihs @thpreihs]

@rafaelgloves @luanalopeslara
"You miss 100% of the shots will don't make" 🤣🤣🤣

[Thiago Preihs @thpreihs]

@rafaelgloves Era meu primeiro projeto como empreendedor... muitos erros ao longo do caminho (que acabei abandonando) mas certamente se tivesse nos EUA nessa época eu conseguiria ir mais longe! Parabéns a Luana e ao time da Kalshi!!!

[Victor Cioffi 🇧🇷 (Neobanks/acc arc) @cioffi_victor]

Eduardo Saverin, Henrique Dubugras, Pedro Franceschi, Luana Lopes Lara… and more

The Brazilian Storm ⛈️🇧🇷

[Dudes Posting Their W’s @DudespostingWs]

This is the kind of good thats so good that people who dont know whats good think its bad

[Jared Carrabis @Jared_Carrabis]

Red Sox 3, Yankees 1

[sasha @lobotoshasa]

Confere, meninos ?

[Thiago Preihs @thpreihs]

@leadlagreport Real deal

[unusual_whales @unusual_whales]

"Panic selling in the stock market is starting to pick up," per CNBC