Activity
-
Can KV cache optimization go beyond just reducing memory footprint? Our latest work, SwiftKV, does exactly that—cutting prefill computation by half…
Can KV cache optimization go beyond just reducing memory footprint? Our latest work, SwiftKV, does exactly that—cutting prefill computation by half…
Liked by Adam Zhao
-
From the re:invent keynote stage More than 50% of new CPU capacity landed is #Graviton
From the re:invent keynote stage More than 50% of new CPU capacity landed is #Graviton
Liked by Adam Zhao
-
🚀 Get Ready for a bunch of exciting announcements that our team has been working for a long time! #AWSreInvent 2024! Over the coming days, I'll be…
🚀 Get Ready for a bunch of exciting announcements that our team has been working for a long time! #AWSreInvent 2024! Over the coming days, I'll be…
Liked by Adam Zhao
Experience
Education
-
Stanford University
-
Artificial intelligence certificate
Courses taken:
CS221 Artificial Intelligence
CS231N Convolutional Neural Networks for Visual Recognition
CS224N Natural Language Processing with Deep Learning
CS330 Deep Multi-Task and Meta Learning -
-
Pioneering OMSCS student. Specialized in Machine Learning
-
-
-
-
Activities and Societies: EECS 320 grader, EECS 311 grader, Ph.D assistant
-
-
Activities and Societies: Student Science and Technology Innovation Association, Student Union, Teaching Assistant of Calculus
More activity by Adam
-
Wow! Customers will love Trn2, Anthropic sees 60% better performance with #trainium2 on Claude 3.5 Haiku and Apple expect 50% better performance with…
Wow! Customers will love Trn2, Anthropic sees 60% better performance with #trainium2 on Claude 3.5 Haiku and Apple expect 50% better performance with…
Liked by Adam Zhao
-
A good opportunity to learn everything about Trainium2!
A good opportunity to learn everything about Trainium2!
Shared by Adam Zhao
-
Two of my favorite people speaking about one of my favorite technologies, you don't want to miss it!
Two of my favorite people speaking about one of my favorite technologies, you don't want to miss it!
Liked by Adam Zhao
-
Are you ready to conquer the computational demands of Generative AI? 🚀 I will join Gadi Hutt's session and talk about how we optimize and scale…
Are you ready to conquer the computational demands of Generative AI? 🚀 I will join Gadi Hutt's session and talk about how we optimize and scale…
Liked by Adam Zhao
-
Happy to share that we're expanding our strategic collaboration with Anthropic as their primary training partner in addition to already being their…
Happy to share that we're expanding our strategic collaboration with Anthropic as their primary training partner in addition to already being their…
Liked by Adam Zhao
-
It's Pre:Invent time! Excited to announce that HUGS is now supported and available on Amazon Web Services (AWS) Inferentia2! HUGS or Hugging Face…
It's Pre:Invent time! Excited to announce that HUGS is now supported and available on Amazon Web Services (AWS) Inferentia2! HUGS or Hugging Face…
Liked by Adam Zhao
-
If you are looking for something to read/study this weekend, I added lots of LLM-related bonus from-scratch coding resources over the last few months…
If you are looking for something to read/study this weekend, I added lots of LLM-related bonus from-scratch coding resources over the last few months…
Liked by Adam Zhao
-
I officially started my 2024 Re:Invent journey today and I am super excited about all the GenAI momentum going on here. The in-house ML chips…
I officially started my 2024 Re:Invent journey today and I am super excited about all the GenAI momentum going on here. The in-house ML chips…
Shared by Adam Zhao
-
I'm exciting to share our work on building and optimizing LLM inference at scale, using Amazon's in house chip. "These choices allowed Rufus to…
I'm exciting to share our work on building and optimizing LLM inference at scale, using Amazon's in house chip. "These choices allowed Rufus to…
Liked by Adam Zhao
-
Excited to share what my team’s be working on!
Excited to share what my team’s be working on!
Liked by Adam Zhao
-
We proposed CLIP-MoE, a Mixture-of-Experts version of CLIP, built using a simple yet effective model-agnostic strategy - Diversified Multiplet…
We proposed CLIP-MoE, a Mixture-of-Experts version of CLIP, built using a simple yet effective model-agnostic strategy - Diversified Multiplet…
Liked by Adam Zhao
-
LLM inference latency has become an increasingly important topic. We know how to get good latency for batch=1, and good throughput for larger batch…
LLM inference latency has become an increasingly important topic. We know how to get good latency for batch=1, and good throughput for larger batch…
Liked by Adam Zhao
-
Check out this AWS blog about how Rufus build an LLM application at scale with high scalability, availability, throughput and low latency. It’s a…
Check out this AWS blog about how Rufus build an LLM application at scale with high scalability, availability, throughput and low latency. It’s a…
Shared by Adam Zhao
-
I’m happy to confirm that the physical copies have started shipping. Happy weekend, everyone! P.S. I’ll be at the PyTorch Conference in SF next…
I’m happy to confirm that the physical copies have started shipping. Happy weekend, everyone! P.S. I’ll be at the PyTorch Conference in SF next…
Liked by Adam Zhao
Other similar profiles
Explore collaborative articles
We’re unlocking community knowledge in a new way. Experts add insights directly into each article, started with the help of AI.
Explore MoreOthers named Adam Zhao in United States
-
Adam Zhao
CS and Music @ Stanford
-
Adam Zhao
Architecture Student at Syracuse University | Minor in Real Estate
-
Adam Zhao
CEO at Jmax California Co.
-
Adam Zhao
Senior @ Fair Lawn High School | Aspiring Electrical Engineer
27 others named Adam Zhao in United States are on LinkedIn
See others named Adam Zhao