We can't find the internet
Attempting to reconnect
Something went wrong!
Hang in there while we get back on track
Building Practical, Cost-Efficient GenAI Solutions Using Serverless • Veda Raman • GOTO 2023
"Discover how to build practical, cost-efficient GenAI solutions using serverless technologies and services, including Step Functions, foundation models, and SageMaker. Learn how to reduce costs and improve agility in your applications."
- Gen-AI models are used to solve real-world problems and improve applications
- Predictive models are important for use cases, but cost-efficient solutions are needed
- Use of serverless services helps reduce costs
- Step Functions is a low-code, no-code solution for workflow automation
- Foundation models from AI21 Labs, Amazon Bedrock, and Hugging Face can be used
- Prompt engineering is crucial for generating text, especially for generative AI applications
- Bedrock is an API-driven service for accessing large language models
- SageMaker is also a service for machine learning model training and deployment
- Data orchestration and choreography are important for managing data in serverless workflows
- Cost optimization is crucial for building cost-efficient Gen-AI solutions
- Serverless services can help reduce costs and improve agility in building applications
- Generative AI applications require both deterministic and probabilistic models