Scott Perry
Employers on the program
- 2022–2025Amazon Web Services3 talks
The year around now
Talks in the past 12 months and announced talks ahead, one row per conference, January to December. Hover a mark for the talk.
Appearances by year
- AWS re:Invent
Talks
| Date | Event | Conference | Role | Title | Links |
|---|---|---|---|---|---|
| Dec 1, 2025 | AWS re:Invent 2025 | AWS re:Invent | code talk | Performance engineering on Neuron: How to optimize your LLM with NKI as AWS Trying to eek out every ounce of performance of your LLM? Trying to speed up inference or understand what is going on inside a language model? | Video |
| Dec 2, 2024 | AWS re:Invent 2024 | AWS re:Invent | builders session | Fine-tune Hugging Face LLMs using Amazon SageMaker and AWS Trainium [REPEAT] as AWS Large language models (LLMs) are pretrained on vast amounts of data and perform well across a variety of general-purpose tasks and benchmarks without further…Show lessLarge language models (LLMs) are pretrained on vast amounts of data and perform well across a variety of general-purpose tasks and benchmarks without further specialized training. | |
| Nov 27, 2023 | AWS re:Invent 2023 | AWS re:Invent | code talk | Deploy LLMs with AWS Inferentia & Ray to optimize performance and cost [REPEAT]Generative AI and large language models (LLMs) inspired many organizations to reimagine the experiences they are building for their customers. As these…Show lessGenerative AI and large language models (LLMs) inspired many organizations to reimagine the experiences they are building for their customers. As these sophisticated LLMs are integrated into more applications, developers are challenged with model serving for high-volume deployments while meeting performance targets. | |
| Nov 27, 2023 | AWS re:Invent 2023 | AWS re:Invent | builders session | Fine-tune Hugging Face LLMs using Amazon SageMaker and AWS TrainiumLarge language models (LLMs) are pretrained on vast amounts of data and perform well across a variety of general-purpose tasks and benchmarks without further…Show lessLarge language models (LLMs) are pretrained on vast amounts of data and perform well across a variety of general-purpose tasks and benchmarks without further specialized training. | |
| Nov 28, 2022 | AWS re:Invent 2022 | AWS re:Invent | chalk talk | AWS managed services to run ML training workloads at scale with AWS Trainium as AWS Amazon EC2 Trn1 instances, powered by AWS Trainium chips, are purpose-built for high-performance, deep-learning training and offer up to 50 percent…Show lessAmazon EC2 Trn1 instances, powered by AWS Trainium chips, are purpose-built for high-performance, deep-learning training and offer up to 50 percent cost-to-train savings over equivalent GPU-based instances. Attend this chalk talk to hear about AWS managed services you can use to run your machine learning (ML) training workloads at scale on Trn1 instances. | Slides |
Frequent co-speakers
| Speaker | Shared events | Shared talks | Last together |
|---|---|---|---|
| Matthew McClean | 4 | 1 | 2025 |
| Aaron Sempf | 4 | — | 2025 |
| Abbas Mooraj | 4 | — | 2025 |
| Abhishek Srivastav | 4 | — | 2025 |
| Adam Hill | 4 | — | 2025 |
| Aditya Samant | 4 | — | 2025 |
| Adnan Bilwani | 4 | — | 2025 |
| Ahuva Hazan-Fuchs | 4 | — | 2025 |
| Alan Halachmi | 4 | — | 2025 |
| Alex DeBrie | 4 | — | 2025 |
| Alexander Dragunov | 4 | — | 2025 |
| Alexandra Huides | 4 | — | 2025 |
| Alexis Marcou | 4 | — | 2025 |
| Ali Alemi | 4 | — | 2025 |
| Ali Saidi | 4 | — | 2025 |
Speakers with the same name are merged into one entry, so this page may combine more than one person. Affiliations are shown as each conference listed them.