
Wannabe engineer @ITBA Construyendo https://nitter.cf/t.co/hwsjVKvkJl Ex director de Tech Trek
Buenos Aires, Argentina
Joined June 2018
- Tweets732
- Following434
- Followers379
- Likes30.2K
Eduardo Tormakh retweeted
mi plan de domingo gracias @Keepanalog_ 🫶🫶🫶
Eduardo Tormakh retweeted
Espectacular primera impresión de @Keepanalog_ !!
Gracias @edutormakh increíble producto
Eduardo Tormakh retweeted
Hace rato vengo armando algo para los hinchas de Boca, y qué mejor día que hoy para lanzarlo.
Mi BombonERA: un archivo con estadísticas de partidos, jugadores, técnicos y rivales desde 1905.
Y además, podés guardar tu historial de partidos en cancha.
Entrá y cargá los tuyos 👇
must watch...
A year ago @slashapp had zero posts on socials & no marketing team.
Today we’re at 20M+ impressions & tens of thousands of followers.
Up until now the entire content team has been 2 people.
This is how I've used AI to make this possible:
5
361
Eduardo Tormakh retweeted
pasamos los ultimos meses probando apps de real estate de todo el mundo para diseñar la app de roomix.
va a ser mucho mas inteligente que la web, con nuevas funcionalidades y una experiencia completamente distinta.
anotense en la waitlist para ser beta testers
roomix.ai/app
Soy el unico que prefiere que le aumenten el precio al producto y que me digan "Envio gratis"?
Como puede ser que por enviarme a una misma direccion (en CABA) me cobren mas del doble entre proveedores???
9
338
Solo hay una manera de terminar el día en el que se retira el 10...
youtu.be/8zHypvOmghE?si=WaXo…
8
597
Lo que daria para ver a @marcos_galperin y @HernanK en esto.
2
12
442
Giving off same vibes...
Replying to @v0xium
You need to understand that none of this has any value in a couple years and you'd be better off shoveling pig shit on a farm. Even though software isn't even close to being solved today, the clever architecture that solves all of deep learning's problems is near. When the actual singularity hits, a new world is starting based around exploration. We are rewinding back to the age of discovery the day we discovered fire. The world will be a jungle. How much you travel or have traveled will be the new defining characteristic of human worth, both physical and astral travels. Your goal post-singularity is to bring back novel experiential data to the planetary learning cluster, Gaia. Become a musician, become a new person, don't expect any of this to stick around the same way that our own jobs didn't stick around. AI will also "lose its job" if that makes sense, and that will happen in a flash. Each phase is temporary, and the phases are getting shorter and shorter. Any regulation being designed around today's AI is a massive mistake. None of this will last. This isn't even 1% of the actual singularity. There is much more beyond deep learning of today. This is a plateau as we figure out the next architecture that multiplies everything we already have. This isn't the destination, not even close. It will solve physics. It will solve everything. We might have the power of God over matter, the quantum mechanical 3D printer of matter, a matter compiler that can materialize any state of matter by splicing atoms directly like CRISPR. We haven't even started yet. Everything could change. The fantasy wizard wonderland may happen. Nothing is certain yet. This entire world of today could be a stupid dream that we all awaken from. The idea of a "job" will seem retarded in retrospect, like what exactly were we attached to?
6
517
Eduardo Tormakh retweeted
We bought an accounting firm and took its operating profit margin from 5% to 70%.
We rebuilt the entire firm to be AI native. Now we're partnering with firms across the country to do this at scale. With AI, we can deliver accounting services faster, cheaper, and more accurately to millions of small businesses that have traditionally seen clean books as a luxury.
Humans are not going anywhere. Accountability, empathy, and human context are still critical in this business, and staff will simply shift to higher value services like advisory work while an army of agents handles routine tasks.
We're currently both acquiring accounting firms and working with firm owners to make their operations AI native. If you're a firm owner, please reach out.
Eduardo Tormakh retweeted
We are reseting usage for all paid users of Codex and ChatGPT Work.
Please continue reading for an update on Codex usage limits. The team has been working around the clock, going through thousands of reports and shipping fixes.
Depending on how you use Codex, you should see your usage go between 10% and 50% further than before.
We really went with a fine comb, with many uncovered small things being longstanding and here is what we found and fixed:
- Compaction. We were keeping old images during compaction, sometimes making the context large enough to trigger compaction again. After the fix, usage dropped around 10% for users making heavy use of images. Fixed.
- Memory. Background memory workers could inherit Stop hooks and keep running when the hook wouldn’t let them finish. This affected fewer than 1% of users, with the long tail being pretty bad and we saw one example thread check whether it could stop 15,000 times. Fixed.
- Goals. In some cases, a set /goal could finish and then keep going past the intended stop condition, or the model would keep retrying broken tools without stopping. We saw examples consume anywhere from 15% to 70% of a weekly allowance. Fixed.
- Automations. Some custom schedules could run more frequently than configured. Fixed.
- Subagents. Smaller models (e.g. Luna) sometimes picked more capable helpers without being explicitly asked. The same was true where the orchestrating model not running in /fast mode could request sub-agents to run /fast. Fixed.
- Computer History. The older implementation could lead to repeatedly summarizing overlapping activity. For some cases we saw it consume up to one fifth of the weekly usage per week. Fixed.
- Rolling task summaries. Ordinary turns were triggering extra background requests. These added about 1% to token usage. Small each time, but it adds up. We have disabled this.
- MCP. Some tool results could be encoded twice. We also found tool instructions getting cut off and fetched again. Fixed.
We’ve also made architectural changes to prevent these from regressing and our teams will get paged if it happens regardless. We are also working on showing you directly in the app where your usage goes so you don’t have to guess.
Goes without saying that we’re resetting usage limits and I hope you enjoy a very nice Saturday!
Eduardo Tormakh retweeted
We’re sorry to see that OpenAI put out a note saying they plan to block Cursor users from accessing OpenAI models in three months.
OpenAI models serve about 5% of Cursor user traffic, and we’re speaking with the OpenAI team to resolve this.
Cursor was one of the very first users of OpenAI, we’ve worked closely with their team for years, and we’ve trusted their platform to be neutral infrastructure for our business.



















