Andy Lee
@andylizf
Followers
21
Following
63
Media
0
Statuses
6
Joined July 2022
Happy to share our lab's new work, "Barbarians at the Gate"! I contributed to the 'Can't Be Late' case study, exploring how to apply AI to automatically generate scheduling policies for deadline-sensitive jobs. Honored to collaborate with this great team. Comments welcome! 🥳
🚀 Excited to release our new paper: “Barbarians at the Gate: How AI is Upending Systems Research” We show how AI-Driven Research for Systems (ADRS) can rediscover or outperform human-designed algorithms across cloud scheduling, MoE expert load balancing, LLM-SQL optimization,
0
1
6
🚀 3 days to go! The SkyPilot AI Infra Meetup is this Thursday! Talks from @nvidia, @phonic_co, co-hosted with @nebiusai. Come learn how to build multi-cloud AI infra, get free cloud credits and meet the SkyPilot community! https://t.co/ee9XF6Hzw5
luma.com
Join us for the SkyPilot AI Infrastructure Meetup with Nebius! We'll be talking about open-source AI infra, using Kubernetes for AI, building AI platforms on…
0
6
13
The future of AI is personal and private. LEANN is a foundational step in that direction.Honored to have contributed to this vision with the incredible team at @BerkeleySky. The thread below has all the details. #OnDeviceAI #FutureOfAI
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: https://t.co/QwkYx1t0oa 📄
0
0
2
VeRL now officially supports launching via SkyPilot! Let SkyPilot deal with infra heavylifting for @verl_project: 🚀Spin up VeRL workers on your k8s or clouds 🔧Set up ray 🤖Ignite your agentic RL training Check out the VeRL doc: https://t.co/o7AzrEMq9o
1
5
17
LEANN: A Low-Storage Vector Index @YichuanM et al. introduce a storage-efficient ANN search index that reduces storage to under 5% of raw data while maintaining 90% top-3 recall in under 2 seconds through graph-based recomputation and pruning. 📝 https://t.co/JrTnv7ENnE
arxiv.org
Embedding-based search is widely used in applications such as recommendation and retrieval-augmented generation (RAG). Recently, there is a growing demand to support these capabilities over...
0
5
13