Ben Geist
666 posts
Ben Geist
@b_geist
MLE @ramplabs / physics + math nerd / Kate Bush fan
Brooklyn, NY Katılım Temmuz 2019
330 Takip Edilen222 Takipçiler
@realmcore_ Have yall experimented with that ? Been doing some work on pulling context from a large context corpus (remembering) and selectively pruning intra run and seeing some promising results
English
RLMs seem like a precursor to effective forgetting mechanisms. LLMs are stateless, and so an RLM can be seen as an LLM that selectively discards context at the end of a task. In the future, I suspect memory will be allocated dynamically and efficiently, not rule based
akira@realmcore_
English
Friend of mine (@4pranavc) just launched a super cool product to finetune models in a really easy way
Provide the model, dataset, choose an algo (SFT, GRPO, etc.) -> run the job, pretty neat!
If you’re doing any finetuning, worth checking out:
vibetune.framer.ai
English
It's over. Time to check on your portfolio.
Drop your brokerage CSV into Ramp Sheets. It'll calculate your real P&L, flag overexposure, and help you decide what to buy, hold, or sell.
Citrini@Citrini7
JUNE 2028. The S&P is down 38% from its highs. Unemployment just printed 10.2%. Private credit is unraveling. Prime mortgages are cracking. AI didn’t disappoint. It exceeded every expectation. What happened? citriniresearch.com/p/2028gic
English
Team recorded this yesterday from the Super Bowl
Ramp Labs@RampLabs
Ramp Sheets now works on mobile. Run financial models and ask questions from your phone, today.
English
A finance team just sent us this. They ran a full budget vs. actuals analysis in under 5 minutes.
Uploaded their budget P&L and 10 months of GL actuals. Ramp Sheets matched every line item, calculated variances, and flagged problem areas.
This used to take half a day.
English
@paulscherer I like @RampLabs building a financial model every time there's a big deal announced. Do I think people will ditch Excel for this? No. Does it make me think of Ramp? Yes! x.com/RampLabs/statu…
Ramp Labs@RampLabs
xAI just raised $20B at ~$230B. Ramp Sheets thinks it's a good deal: $383/user vs OpenAI's $1,875/user (80% discount) 600M users who already open X daily Only need 8% converting to $20/mo Grok to justify valuation
English
@zachmoskow @eglyman You can play around with the sheet and assumptions here: labs.ramp.com/sheets?share=8…
English
@eglyman I think the research showed they’re at like 1/10th of that MAUs. How does rerunning it on that work?
English
In 2026, may you have the confidence of an AI model that just analyzed its own company and concluded it’s deeply undervalued
Ramp Labs@RampLabs
xAI just raised $20B at ~$230B. Ramp Sheets thinks it's a good deal: $383/user vs OpenAI's $1,875/user (80% discount) 600M users who already open X daily Only need 8% converting to $20/mo Grok to justify valuation
English