Prototyping Engineer | Amazon Web Services
Exploring modern applications with opensource large language models
Generative AI is a rapidly developing field that has the potential to transform the way we interact with computers. Generative AI models can create new content that is both realistic and creative by learning from large datasets of text, images, and code. This opens up the possibility of developing a wide range of applications for a variety of business use cases.
In this session, we will look at some of the solutions that use open source large models and customisation options to solve some of the most common use cases such as question and answer, summarisation, simplification, tone augmentation, and so on. we will briefly explore advanced prompt engineering, fine tuning and RLHF options to improve the results of the open source models and see a quick demo of one of the patterns at the end of the session.
TRACK: AI & ML
19 Oct 2023 | Time: 12:00-01:00 PM
Aastha Varma is a Prototyping Engineer for the India Prototyping Team at AWS. Inspired and passionate about social good, open source and technology. Works with diverse AWS customers from Startups to Enterprises, helps contextualize complex use cases in AIML to GenAI, architect solutions and build innovative-impactful prototypes. Currently – learning, experimenting and implementing solutions to build Generative AI applications for customers. Use-cases range from evaluating LLMs, cost efficient finetuning, and benchmarking inferences to get best price-performance.