anu
@sigdel29
Followers
206
Following
545
Media
88
Statuses
2K
Debugging reality, one commit at a time,
New York
Joined May 2025
LongMem exposes a fundamental problem with LLMs: just because information is in the context window doesn't mean the model can use it. LongMem tests memory recall by asking the same questions under two conditions. With Claude Haiku 4.5 as baseline: * Minimal context (just the
3
6
17
sam buddy did u not hear? memory is a reasoning task !!!
📁 Sam Altman says the real AI breakthrough will not be better reasoning, but total memory. An AI will be able to remember every conversation, email, and document across a person’s lifetime, identifying patterns and preferences humans never consciously express. Once memory
0
1
9
Honcho has fully determined the Pareto Frontier in Agent Memory. It’s state-of-the-art on LoCoMo, LongMem, and BEAM. Honcho is also the fastest, cheapest and most token efficient memory solution.
22
38
130
We posted a ML Engineer role last week. 800+ applicants, only 32 are US Citizens. Is this happening to other founders?
5
2
13
can someone connect me to @OpenAI @OpenAIDevs team to increase rate limits?
0
0
0
REAL STEEL IN NYC!!! This was awesome!!! @REKrobot
NYC Pioneers in humanoid robot combat will be in your town on November 25th. History is being made! 🤖🦾🦿⚙️🪛🔩💥🥊
1
0
5
Hired our first teammate a month ago. About to make an offer to another today. We’re hiring just 1 more founding engineer… Are you an ML Engineer??? dm me⚡️
2
4
11
when i was growing up, i was an apple fan to the core- to a point i used to run an iphone jailbreak side hustle, i had an iphone cake for my birthday i loved the innovation and a lot of the early products really inspired me to be in tech, its sad that's not the case anymore
When a company releases something that is so obviously underwhelming then the natural question is: “Did no one at the company see how bad this is OR did no one have the courage to speak up?” I’m not sure which is worse.
0
0
0
today, we're releasing the largest egocentric dataset of physical jobs - 400k action labels - 2.5k clips - 2x'd open source dataset size (download below)
113
189
2K
i'm gonna quote this to my non tech friends so hard
unexpectedly found @SadhguruJV to be a strong AI accelerationist "AI is liberation from mental labor, free yourself from being a data coolie"
0
0
0