Sarah Wooders (@sarahwooders) / X

Sarah Wooders

1,270 posts

Sarah Wooders

@sarahwooders

co-founder & CTO @Letta_AI making machines that learn prev phd @ucbrise @BerkeleySky + @MIT

San Francisco, CA

Joined May 2010

Sarah Wooders
@sarahwooders
Sep 23, 2024
Excited to announce @Letta_AI, the company @charlespacker and I started for building stateful LLM agents We're building out an incredible (in-person) team in SF, and are actively hiring founding engineer/researchers jobs.ashbyhq.com/letta techcrunch.com/2024/09/23/let…
jobs.ashbyhq.com
Letta Jobs
Letta Jobs
65K
Sarah Wooders
@sarahwooders
Nov 4, 2025
the hardest part about being a female founder.. in my experience is finding a boyfriend / partner that understands the short term gains that we are forgoing and the long-term optimization game that we are playing. most men in SF (as we women known) insist on paying for dates,
Jeson Lee
@thejesonlee
Nov 3, 2025
the hardest part about being a male founder.. in my experience is finding a girlfriend / partner that understands the short term gains that we are forgoing and the long-term optimization game that we are playing. most women expect to have most dates paid for, their vacations
51K
Sarah Wooders
@sarahwooders
Nov 7, 2024
Excited to release the first course on agents with memory with @AndrewYNg @DeepLearningAI @Letta_AI We cover customizing agent memory management beyond MemGPT, integrating external tools @langchain @composio @crewAIInc, agentic RAG with @llama_index, and multi-agent.
Andrew Ng
@AndrewYNg
Nov 7, 2024
New short course: LLMs as Operating Systems: Agent Memory, created with @Letta_AI, and taught by its founders @charlespacker and @sarahwooders. An LLM's input context window has limited space. Using a longer input context also costs more and results in slower processing. So,
00:00
13K
Sarah Wooders
@sarahwooders
Sep 5, 2025
We've been trying to wrangle Responses API, and here's my response to this: Claim #1 -> Responses is just ChatCompletitions++ Wrong! Responses makes managing your own context window a PINA and also introduces a brand new (not generally supported) API that is very obviously
prashant
@prashantmital
Sep 5, 2025
feeling frustrated (and a little guilty): there’s still way too much confusion about @OpenAI's Responses API. this is partly on us: we haven’t always been clear about why we built it, how to use it, and why it matters. here's my attempt at setting the record straight. 👇
14K
Sarah Wooders
@sarahwooders
Oct 28, 2025
Pretty crazy to see Anthropic API being implemented over ChatCompletions, which used to be the standard (until OpenAI refused to properly support reasoning)
Skyler Miao
@SkylerMiao7
Oct 28, 2025
Hey guys, thank you all for your love and passion for M2. Lot asking why we recommend Anthropic API. I think need to explain a little bit. M2 is a agentic thinking model, it do interleaved thinking like sonnet 4.5, which means every response will contain its thought content. Its
24K
Sarah Wooders
@sarahwooders
Feb 6, 2025
We are moving from a stateless to stateful LLM programming paradigm: from “agents” more akin to stateless workflows, to agents that can actually learn, evolve, and persist over time.
11K
Sarah Wooders
@sarahwooders
Feb 18, 2025
All I want is a model that - can properly call tools - API support - does what I tell it to lmk when there's an alternative to claude 3.5 sonnet
7.7K
Sarah Wooders
@sarahwooders
Sep 26, 2024
Most "AI danger" stuff is just marketing to portray models are more powerful than they actually are imo. Now we're in a ridiculous situation where the foundation model marketers + AI doomers have teamed up to use the government to try to stifle OSS innovation. Thank you
Andrew Ng
@AndrewYNg
Sep 26, 2024
A decision on SB-1047 is due soon. Governor @GavinNewsom has said he's concerned about its "chilling effect, particularly in the open source community". He's right, and I hope he will veto this. If you agree, please like/retweet this to show your support for VETOing SB-1047!
11K
Sarah Wooders
@sarahwooders
Feb 12, 2025
Sounds like OpenAI wants to make it harder to directly access models via APIs - which means more black box systems that are hard to debug and understand. This is why we need open models and model providers who aren't obsessed with hiding everything from developers.
Sam Altman
@sama
Feb 12, 2025
OPENAI ROADMAP UPDATE FOR GPT-4.5 and GPT-5: We want to do a better job of sharing our intended roadmap, and a much better job simplifying our product offerings. We want AI to “just work” for you; we realize how complicated our model and product offerings have gotten. We hate
2.5K
Sarah Wooders
@sarahwooders
Aug 12, 2025
A few months ago, @mem0ai published benchmarking numbers for MemGPT and claimed "SOTA" in memory. The weird part? I have no idea how they even could have run that benchmark, without making significant modifications to MemGPT (they did not respond to our requests for details on
Letta
@Letta_AI
Aug 12, 2025
Simply storing the conversation history in a file outperforms specialized tools for agent memory on LoCoMo, a popular retrieval benchmark (74.0% versus reported 68.5%)
6.3K
Sarah Wooders
@sarahwooders
Oct 22, 2025
Legacy??
3.7K
Sarah Wooders
@sarahwooders
Aug 4, 2024
Office setup day with @charlespacker
3.6K
Sarah Wooders
@sarahwooders
Nov 4, 2025
Replying to @zakkinion
Have fun at IHOP 👋
1.3K
Sarah Wooders
@sarahwooders
Oct 13, 2025
Don't let LLM providers own your data - separate agent storage from compute with @Letta_AI
Eric Hartford
@QuixiAI
Oct 13, 2025
My account was completely deleted. My appeal was denied with no consideration and no process. all my history is gone. no chance to export and save my data. I am completely innocent of any wrongdoing. I have absolutely no idea why @sama would do this to me. @OpenAI
4.7K