AWS Annapurna Labs - Micro 2024 Conference Coffee Break

Date

Start

End

Interested in meeting with the team?
 
Check into the Coffee Break via the "Register" button above to register your information to be contacted for potential career opportunities with AWS Annapurna Labs.
 
To view and apply for our current openings, click here.
 
Can't find the perfect fit? Remember to upload your resume via the "Register" link below and a member of our team will be in touch.

About us. As a fabless chip start-up in 2011, we focused on bringing innovation to the fast-growing cloud infrastructure. Four years after inception, Annapurna Labs was acquired by Amazon Web Services (AWS). Since then, we have accelerated its innovation and developed a number of products that benefit cloud customers, including AWS Nitro technology, Inferentia custom Machine Learning chips, and AWS Graviton2 processors. Today, we're building the most performant, cost-effective Generative AI insfrastructure and software on the planet.

 

 

 

AWS Neuron

AWS Neuron is an SDK with a compiler, runtime, and profiling tools. With Neuron, you can use popular frameworks, such as TensorFlow and PyTorch, and optimally train and deploy machine learning models on Amazon EC2 Trn1 and Inf1 instances with minimal code changes and without tie-in to vendor-specific solutions.

Check it out.

AWS Trainium

AWS Trainium is the second-generation machine learning (ML) chip that AWS purpose built for high-performance, cost-effective training of generative AI models. Trainium has been optimized for training natural language processing, computer vision, and recommender models used in a broad set of applications, such as speech recognition, recommendation, fraud detection, image recognition, and forecasting.

Check it out.

AWS Inferentia

AWS Inferentia is Amazon's first custom silicon designed to accelerate deep learning workloads and is part of a long-term strategy to deliver on this vision. Customers can use Inf1 instances to run large scale machine learning inference applications such as search, recommendation engines, computer vision, speech recognition, natural language processing, personalization, and fraud detection.
 
AWS Graviton
 
 
 
AWS Graviton processors offer optimal cost-effective performance for cloud workloads on Amazon EC2, with Graviton2 processors presenting a significant performance boost over the first generation. Graviton2-based instances cater to diverse workloads including application servers, machine learning inference, and gaming, supported by various AWS services. Graviton3 processors, the latest iteration, offer up to 25% better compute performance, double floating-point and cryptographic performance of Graviton2, along with support for DDR5 memory. Graviton3E processors further enhance vector-instruction performance by up to 35%, particularly benefitting high-performance computing applications. 

Check it out.
In the news
 
  • Inside the AI chip race: How a pivotal happy hour changed Amazon's strategy in the cloud (GeekWire, November 2023)

  • About AWS Generative AI Chips (YouTube, June 2023)

  • What you need to know about the AWS AI chips powering Amazon's partnership with Anthropic (October 2023)

  • Andy Jassy's Annual Letter to Shareholders (April 2024)
    "...Sometimes, people ask us “what’s your next pillar? You have Marketplace, Prime, and AWS, what’s next?” This, of course, is a thought-provoking question. However, a question people never ask, and might be even more interesting is what’s the next set of primitives you’re building that enables breakthrough customer experiences? If you asked me today, I’d lead with Generative AI (“GenAI”). 

    Customers have asked us to push the envelope on price-performance for AI chips, just as we have with Graviton for generalized CPU chips. As a result, we’ve built custom AI training chips (named Trainium) and inference chips (named Inferentia). In 2023, we announced second versions of our Trainium and Inferentia chips, which are both meaningfully more price-performant than their first versions and other alternatives. This past fall, leading FM-maker, Anthropic, announced it would use Trainium and Inferentia to build, train, and deploy its future FMs. We already have several customers using our AI chips, including Anthropic (think: Claude3), Airbnb, Hugging Face, Qualtrics, Ricoh, and Snap.

Date

Start

End