Eric Zelikman

1.2K posts

Eric Zelikman banner
Eric Zelikman

Eric Zelikman

@ericzelikman

cofounder & ceo @humansand - building ai for humans // was lgtm-ing @xAI, phd-ing @stanford

Katılım Nisan 2010
2K Takip Edilen23.7K Takipçiler
Sabitlenmiş Tweet
Eric Zelikman
Eric Zelikman@ericzelikman·
finally announcing i’ve started humans& w/ amazing friends @gharik & @YuchenHe07 & @TheAndiPenguin & @noahdgoodman & many other world-class folks. we're optimists: it’s possible to rethink how we build ai, to empower people to accomplish more together tldr: love is all you need
humans&@humansand

Today we introduce humans&, a human-centric frontier AI lab. We believe AI can be reimagined, centering around people and their relationships with each other. At its best, AI should serve as a deeper connective tissue that strengthens organizations and communities

English
136
40
673
124.3K
Broskі
Broskі@broskiFGC·
its cool how half the global economy is contingent on this shit
Broskі tweet media
English
193
1.2K
32.5K
2.2M
Eric Zelikman
Eric Zelikman@ericzelikman·
🚀 the @radixark team has already contributed a lot to open source infra - excited to see what they do next
RadixArk@radixark

Today, we are thrilled to officially launch RadixArk with $100M in Seed funding at a $400M valuation. The round was led by @Accel and co-led by @sparkcapital. RadixArk exists to make frontier AI infrastructure open and accessible to everyone. Today, the systems behind the most capable AI models are concentrated in a small number of companies. As a result, most AI teams are forced to rebuild training and inference stacks from scratch, duplicating the same infrastructure work instead of focusing on new models, products, and ideas. RadixArk was founded to change that. We are building an AI platform that makes it easier for teams to train and serve the best models at scale. RadixArk comes from the open-source community. We started with SGLang, where many of us are core developers and maintainers, and expanded our work to Miles for large-scale RL and post-training. We will continue contributing to both projects and working with the community to make them the strongest open-source infrastructure foundations for frontier AI. We would like to thank our long-term partners, contributors, and the broader SGLang community for believing in this mission. We're also grateful to @Accel and @sparkcapital, NVentures (Venture capital arm of @nvidia), Salience Capital, A&E Investment, @HOFCapital, @walden_catalyst, @AMD, LDVP, WTT Fubon Family, @MediaTek, Vocal Ventures, @Sky9Capital and our angel investors @ibab, @LipBuTan1, Hock Tan, @johnschulman2, @soumithchintala, @lilianweng, @oliveur, @Thom_Wolf, @LiamFedus, @robertnishihara, @ericzelikman, @OfficialLoganK, and @multiply_matrix among others. Thanks for the exclusive interview with @MeghanBobrowsky at @WSJ about our vision.

English
3
4
44
4K
Eric Zelikman retweetledi
LMSYS Org
LMSYS Org@lmsysorg·
The DeepSeek V4 garbled output bug in open source inference engine is fixed in SGLang. To everyone affected over the weekend, sorry for the trouble. Huge thanks to @Ant_Group for landing the fix PR. It was a cross-company, cross-timezone, sub-48-hour marathon. @ollama and @humansand surfaced it first; @nvidia, @AIatMeta, and @FireworksAI_HQ raised the same signal soon after. @deepseek_ai replied in seconds at every hour. @FireworksAI_HQ stayed up late with us until it shipped. @SemiAnalysis_ and @ollama provided the machines that made the debugging possible. The SGLang team dug in through the weekend. The real OSS is the friends we made along the way.🫶
LMSYS Org tweet media
English
16
27
283
79.1K
Eric Zelikman retweetledi
Ying Sheng
Ying Sheng@ying11231·
Deepseek V4. This is the most comprehensive day 0 support I have ever experienced. Rich SGLang features including hierarchical caching for sparse attention and Miles support for RL. Enjoy 🥰
LMSYS Org@lmsysorg

DeepSeek V4 by @deepseek_ai just dropped! SGLang is ready on Day 0 with a full stack of optimizations from architectures to low-level kernels. We also deliver a verified RL training pipeline in Miles (by @radixark) for V4 at launch: 1️⃣ Native "ShadowRadix" Design: DeepSeek V4's hybrid attention is complex. Our new ShadowRadix engine is the first to provide native prefix caching for SWA and compressed KV pools, making 1M+ context retrieval seamless and memory-efficient. 2️⃣ High-Performance Kernels: - Flash Compressor: IO-aware fused kernels, 10x faster than naive implementations. - Lightning TopK: High-speed indexing for 1M context in just 15µs. - Integrate FlashInfer trtllm-gen MoE, FlashMLA, and MegaMoE kernels 3️⃣ Rich Features: Speculative decoding, HiSparse, Attention DP/TP/CP and MoE TP/EP, and multi-platform support 4️⃣ Verified RL: The open-source RL pipeline: full parallelism (DP/TP/EP/PP/CP), tilelang kernels, tensor-level checked precision, verified with growing reward. Get started immediately with our out-of-the-box Cookbook 👇 Enjoy! #DeepSeekV4 #SGLang #LLM

English
7
27
337
27.2K
Michael Y. Li
Michael Y. Li@michaelyli_·
Can a language model learn, end-to-end, what to keep in its own KV cache and what to throw away? Can it learn to forget while it learns to reason? Deep learning's central lesson: capability emerges from end-to-end optimization, not heuristics/strong inductive biases. But for efficiency, we rely heavily on hand-designed approaches. 🗑️ Introducing Neural Garbage Collection (NGC): we train a language model to jointly reason and manage its own KV cache, using reinforcement learning with outcome-based task reward alone. No SFT, no proxy objectives, no summarization in natural language. New paper with @jubayer_hamid, Emily Fox, and @noahdgoodman!
Michael Y. Li tweet media
English
30
133
905
163.2K
Eric Zelikman
Eric Zelikman@ericzelikman·
@RonConway Wishing you a fast recovery and always incredibly thankful to work with you and the amazing folks at SV Angel 🙏
English
0
0
5
662
Ron Conway
Ron Conway@RonConway·
I want to share some difficult news. I was recently diagnosed with a rare form of cancer and I want you to hear it directly from me. Treatment is starting immediately and will include multiple strategies over the course of about a year. While I will be stepping back from some of my usual activities, I will continue to support SV Angel founders, who I love with a passion. SV Angel remains unchanged. Topher has made all of our investment decisions for the better part of the last decade, and Ronny joined as Managing Partner in 2024. They bring experience from nearly every major technology cycle in Silicon Valley and are now focused on partnering with founders building the future of AI. SV Angel has a deep, experienced team that remains fully focused on supporting exceptional founders. With a more focused and balanced schedule, I can prioritize treatments while helping SV Angel founders at inflection points like we always do! I’ve chosen not to share the specific type of cancer since I don't want speculation about my prognosis. I appreciate your understanding and respect for this. I am optimistic about my prognosis. I am fortunate to have the best/amazing team of UCSF doctors in San Francisco, and as you know, I never back down from a fight. Thank you for your support, it means a great deal to me.
English
593
77
4.3K
467.2K
Joanne Jang
Joanne Jang@joannejang·
i shared this note on slack:
Joanne Jang tweet media
English
288
313
4.8K
1.1M
Guodong Zhang
Guodong Zhang@Guodzh·
Surprised again and again how much a few engineers can do with a clear goal in 3-6 months.
English
7
6
141
13.6K
Eric Zelikman
Eric Zelikman@ericzelikman·
@bryancsk i’ve also wondered why there’s a moon viewing pond in a botanical garden that closes at sunset
English
0
0
6
951
Bryan Cheong
Bryan Cheong@bryancsk·
It is left as an exercise to the reader to figure out why a moon viewing pond should not be covered with duckweed
Bryan Cheong tweet media
English
10
0
74
4.2K
Saurabh Shah
Saurabh Shah@saurabh_shah2·
In hindsight forming my entire perception of this city from Twitter was pretty stupid. It’s really quite lovely here
English
6
1
80
10.5K
Shirley Wu
Shirley Wu@ShirleyYXWu·
To the best PhD years at Stanford To all who have carried my learning and lit the way for my growth ❤️ Thank you is hardly enough
Shirley Wu tweet mediaShirley Wu tweet mediaShirley Wu tweet mediaShirley Wu tweet media
English
47
7
376
32.2K
Eric Zelikman retweetledi
humans&
humans&@humansand·
We will be co-hosting this in SF with our great friends at SV Angel. We'll keep the event small and cozy and have a few prizes for the teams:🥇$16k🥈$8k🥉$4k. Come by and hang out with the humans& team - it’ll be a lot of fun!
humans&@humansand

Announcing the humans& hackathon! Hack with us this Saturday - come experiment and build AI apps to help people collaborate and communicate, work with creative folks, learn a bit about what we're building, and win cool prizes Apply here: luma.com/2pbif8t9

English
2
9
79
20.9K
Eric Zelikman
Eric Zelikman@ericzelikman·
One cost more people (even prospective founders) should consider: if you did something else, would the impact you're having today still happen somewhere? If yes, you're missing the opportunity to change things for the better
Amy Tam@amytam01

x.com/i/article/2023…

English
5
3
138
23.3K
Eric Zelikman retweetledi
humans&
humans&@humansand·
Announcing the humans& hackathon! Hack with us this Saturday - come experiment and build AI apps to help people collaborate and communicate, work with creative folks, learn a bit about what we're building, and win cool prizes Apply here: luma.com/2pbif8t9
humans& tweet media
English
19
23
312
82.4K
Eric Zelikman
Eric Zelikman@ericzelikman·
@EthanChoi7 anthropic’s first round was 124m in 2021. i’d say they’ve done ok since
English
2
0
40
3.4K
Ethan Choi
Ethan Choi@EthanChoi7·
I was trying to think the other day of any success stories of companies that have raised a monster round (100s of $ Ms or $1B+) out of the gate. Anyone have an example or two? 🤔 I had a hard time thinking of many tbh but can think of some blow ups where too much capital pre-PMF wasn’t great like Quibi ($1.75B before launch), WeWork ($12B before firming up of unit economics), etc… One case study might be @anduriltech that raised $127M seed but even other deep tech leaders like @SpaceX raised $20M Series A and @Tesla Series A was $7.5M. Not coming from a place of back-seat driving VC criticism but more of genuine curiosity for case studies… Obviously many stories still early and playing out Many good reasons that capital constraints at the early and mid-stages of a startup create a lot of positive behavior and externalities like scrappiness / efficiency / creativity… However, we are in a new AI and deep tech era where revenue growth and outcomes / possibilities are markedly different so I this approach can definitely work better now than in the prior era. For neo labs like @thinkymachines / @ssi / others, this is the only way to do it given the cost of research talent, $ needed for GPUs / compute for training, and not diluting founders too much so incentives are aligned but will be interesting to see how it plays out…
Garry Tan@garrytan

Big splashy pre-PMF rounds happen on big legible ideas and teams with social proof Don’t hate the player hate the game It doesn’t mean you can’t build a big co if you don’t have the right resume but it does mean you need to be 10x more legible through traction or clarity

English
9
1
54
24.2K
Eric Zelikman
Eric Zelikman@ericzelikman·
We will see people and organizations spend increasingly more time thinking about what to implement and how to implement it than actually implementing it. This will enable but also require new levels of collaboration
will brown@willccbb

with the latest models, i am now finding myself thinking about complex system design problems more, not less. the magnitude of what can be reasonably attempted is monumentally larger. you need to make sure you’re asking it to build the right thing.

English
2
3
64
6.8K

Keşfet

@broskiFGC @radixark @Ant_Group @ollama @humansand @nvidia @AIatMeta @FireworksAI_HQ