Joey Gonzalez
@profjoeyg
Followers
4K
Following
940
Media
46
Statuses
684
Professor @UCBerkeley, co-director of @LMSysorg, and co-founder @RunLLM
Berkeley, CA
Joined June 2011
RT @RunLLM: 🗓️ ICYMI. Run of the Week | Aug 31, 2025.This week at RunLLM: blogs, features, and hot takes worth a second look. 📌 The AI Fro….
0
2
0
RT @shiyi_c98: An amazing collaboration with the Biomni team! We introduce Biomni-R0, a reasoning-enabled multi-task multi-tool biomedical….
0
9
0
RT @adityagp: New research agenda we're kickstarting at Berkeley: redesigning data systems to serve the dominant workload of the future: ag….
0
41
0
RT @YichuanM: Thanks for sharing! We’re actively building LEANN, come try it out and share your feature ideas. We’re excited to make it ful….
0
3
0
RT @Saboo_Shubham_: RAG everything locally!. Index millions of documents on your laptop using 97% less storage than traditional vector data….
0
238
0
RT @vsreekanti: Whether you like GPT-5 as a consumer or not, its usage in AI applications is distinctly different from previous flagship mo….
0
3
0
RT @forward_future_: FROM THE LIVE SHOW: @charlespacker, co-founder & CEO of @Letta_AI, on why memory may matter more than the model. “Eve….
0
8
0
RT @lisabdunlap: Don't know what to get your advisor for their birthday? Give them the best gift of all: their time back. Happy (late) bi….
0
8
0
RT @vsreekanti: Once we all have swarms of agents working for us, we're going to want those agents to work together. That means we need a….
0
2
0
RT @sreekanti: @vsreekanti @profjoeyg Humans collaborate efficiently on Slack. Agents should too. We already have the message bus, why inve….
0
1
0
We need a good benchmark for memory agents.
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.
0
2
11
RT @berkeley_ai: A feature of BAIR faculty @istoica05.with quotes and mentions from fellow BAIR faculty.@profjoeyg, @matei_zaharia, @jennif….
0
6
0
I am really excited to announce the release of LEANN. Most vector stores more than double the size of the data they index. With LEANN, we only increase storage costs by a few percent. We made it easy to use LEANN to index your email and document folders and apply RAG without.
1/N 🚀 Launching LEANN — the tiniest vector index on Earth!. Fast, accurate, and 100% private RAG on your MacBook. 0% internet. 97% smaller. Semantic search on everything. Your personal Jarvis, ready to dive into your emails, chats, and more. 🔗 Code: 📄
0
5
33
RT @AstasiaMyers: Today, crafting agent workflows is hard and technical. In a few years, it’ll be as trivial. A startup's edge won't be it….
0
1
0
RT @HannaHajishirzi: Congratulations to @sewon__min for winning yet another dissertation award, this time, ACL dissertation award. So proud….
0
16
0
RT @Letta_AI: Did you know you can copy content from our documentation to a markdown string? . Great for pasting into Claude Code. https://….
0
1
0
RT @RunLLM: One of the most powerful features in RunLLM v2 is now live:. ✨ Agentic Reasoning with real log + telemetry analysis. 🎥 Watch….
0
3
0
RT @TheHouseVC: Real AI ROI is rare, and @RunLLM is delivering it at scale. With today’s v2 launch, they’re now deflecting 99% of support….
0
2
0
RT @vsreekanti: Our original UI for @RunLLM was shaped by 2 years of constant iteration. And while it worked, it wasn't built for what cust….
0
2
0