Sky Computing
Towards Utility Computing for the Cloud
News
May 11, 2026
Frontier-CS Goes Live: 2,000 Humans vs. AI on an Open-Ended Problem
We placed an open-ended optimization problem in CALICO, UC Berkeley’s official programming contest. To score, you had to beat a frontier AI. Out of 2,000+ contestants, only one submission surpassed the strongest AI agent. CALICO (CALICO Informatics Competition) is the official competitive programming contest hosted by a group of UC Berkeley students interested in competitive coding at the end of each semester. The contest is co-organized with ICPC@Berkeley and Upsilon Pi Epsilon (UPE), a UC Berkeley EECS student organization affiliated with the EECS department. The contest runs on a custom judge platform built on top of DOMjudge, the same open-source system used at the ICPC World Finals.
April 8, 2026
Matei Zaharia awarded ACM Prize in Computing for contributions to data systems, enabling AI
Matei Zaharia, associate professor of electrical engineering and computer sciences (EECS) at UC Berkeley, has been awarded the ACM Prize in Computing for his visionary development of distributed data systems and computing infrastructure. In the prize announcement, the Association for Computing Machinery (ACM) noted Zaharia’s development of open-source systems helped enable large-scale machine learning (ML), analytics and AI at a global scale. ACM is the world’s largest educational and scientific computing society, and their ACM Prize in Computing recognizes early-to-mid-career computer scientists whose work has had broad and lasting impact. Recipients receive a $250,000 prize, with financial support provided by an endowment from Infosys Ltd.
April 2, 2026
EECS researchers receive Laude “Slingshot” awards to advance next-generation AI systems
Three research projects featuring contributors from Sky Computing Lab have recently been selected for Laude’s Slingshot program.
March 4, 2026
OpenThoughts-Agent, Continual Learning Benchmark, and MAP announced as Slingshots // TWO projects with Laude Institute
These 14 projects from Stanford, Berkeley, MIT, CMU, UIUC, and Michigan are tackling production deployment, energy constraints, and continual learning. Several are building on infrastructure from Slingshots // ONE. Together, they show what happens when the right researchers get the right resources at the right time: research that ships, gets adopted, and moves the field forward.
November 18, 2025
vLLM is the top open source project on GitHub for 2025
2025’s top projects split between AI infrastructure (vllm, ollama, huggingface/transformers) and enduring ecosystems (vscode, godot, home-assistant).
Events
June 18, 2026
Dissertation Talk: Towards Domain Adaptation, Interpretability, and Deployment in Large Language Models – Vinamra Benara
This talk presents contributions toward domain adaptation, interpretability, and deployment in large language models. I will discuss how retrieval augmented generation and fine tuning can incorporate specialized knowledge, how question-answering can be used to generate more interpretable embeddings …
May 8, 2026
Dissertation Talk: Rethinking Database Optimization for Modern Workloads – Audrey Cheng
Data systems face unprecedented scalability demands as modern applications, especially AI workloads, evolve rapidly. These shifts make it increasingly difficult to maintain both performance and correctness, which are the core properties that databases must provide. In this talk, I discuss how to ret…
May 7, 2026
Dissertation Talk: Algorithmic Methods For Efficient Deep Learning Inference on the Edge – Eyal Sela
Applications such as live video analytics, robotics, and autonomous vehicles require streaming perception to respond rapidly to evolving scenes, while balancing accuracy, latency, and compute cost. In conventional perception, scaling to larger models or higher-resolution inputs can improve accuracy….
May 6, 2026
Dissertation Talk: Building Open Source Inference Serving Systems – Simon Mo
Inference serving has become its own area of systems research. This talk reflects on seven years of open-source work across three layers of the stack: SLO-aware pipeline serving (Clipper, InferLine, Ray Serve), virtual memory for KV caches (vLLM), and GPU kernel multiplexing below the model boundary…
Publications
March 2026
BlendServe: Optimizing Offline Inference with Resource-Aware Batching.
March 2026
Title: An AI Stack: From Scaling AI Workloads to Evaluating LLMs.
February 2026
Quant VideoGen: Auto-Regressive Long Video Generation via 2-Bit KV-Cache Quantization.
February 2026
Qrita: High-performance Top-k and Top-p Algorithm for GPUs using Pivot-based Truncation and Selection.
January 2026
Delta Fair Sharing: Performance Isolation for Multi-Tenant Storage Systems.
January 2026
VisGym: Diverse, Customizable, Scalable Environments for Multimodal Agents.
January 2026
UCCL-EP: Portable Expert-Parallel Communication.
January 2026
Supporting Our AI Overlords: Redesigning Data Systems to be Agent-First.
January 2026
Supporting Our AI Overlords: Redesigning Data Systems to be Agent-First.
January 2026
Text2SQL is Not Enough: Unifying AI and Databases with TAG.
January 2026
SkyNomad: On Using Multi-Region Spot Instances to Minimize AI Batch Job Cost.
December 2025
Speculative Decoding: Performance or Illusion?
Recent Projects
Continual Learning Bench
Sky Computing Story
Berkeley’s computer science division has an ongoing tradition of 5-year collaborative research labs. Recent labs included the AMPLab (ended in 2016) and the RISELab. These labs have had significant impact in both academia and industry. Past labs publish their research at top conferences in systems, databases, and machine learning. On the industrial side, AMPLab and RISELab fostered several successful startups (Databricks, Opaque, Ponder, Anyscale, to name a few). We are excited to announce the Berkeley Sky Computing Lab where we will strike to make cloud computing a true commodity.
Context
The Sky Computing Lab represents the next chapter of data-intensive systems research at Berkeley. Recent years have seen the explosion of cloud computing. Applications are moving their data and computation to the cloud; on-premise services are dying. In doing so, companies have to make difficult choices between the myriad of cloud providers, each with different services or hardware. Lock-in, whether through artificial migration costs, legal constraints or engineering baggage is real. In the Sky Computing Lab, we will leverage distributed systems, programming languages, security, and machine learning to decouple the services that a company wants to implement from the choice of a specific cloud. Much like the Internet today, cloud computing should be an undifferentiated commodity. Applications should run seamlessly on any or multiple clouds.
Mission
Our mission in the Sky Computing Lab is to transform the cloud into an undifferentiated commodity and ease application burden. As in previous labs, we’re all in — working on everything from basic research to software development, all in the Berkeley tradition of open publication and open source software. Our founding team consists of experts in distributed systems, machine learning, security and programming languages. We’ll use this space to lay out our ideas and progress as we go.
Commitment to Diversity
Sky Computing is guided by Berkeley’s Principles of Community and is committed to providing a safe and caring research environment for every member of our community. We believe that a diverse student body, faculty, and staff are essential to the open exchange of ideas that Sky Computing Lab is founded on.
Our head is in the cloud. We are heading for the SKY.