Managing AI Costs with Sonia Cuff
How are you managing your AI costs? Richard chats with Sonia Cuff about how she's been helping sysadmins understand how AI technology is billed out and how to measure them within a given application and across the organization. Sonia starts with some definitions since the term AI is so broad. It's not just about large language models! The conversation also dives into carbon footprints, and using the FinOps models to help with costing for your cloud infrastructure, whether you're using AI or not!
Episode 976 of the RunAs Radio podcast, hosted by Sonia Cuff, Richard Campbell, titled "Managing AI Costs with Sonia Cuff" was published on March 19, 2025 and runs 37 minutes.
March 19, 2025 ·37m · RunAs Radio
Summary
How are you managing your AI costs? Richard chats with Sonia Cuff about how she's been helping sysadmins understand how AI technology is billed out and how to measure them within a given application and across the organization. Sonia starts with some definitions since the term AI is so broad. It's not just about large language models! The conversation also dives into carbon footprints, and using the FinOps models to help with costing for your cloud infrastructure, whether you're using AI or not!
Episode Description
How are you managing your AI costs? Richard chats with Sonia Cuff about how she's been helping sysadmins understand how AI technology is billed out and how to measure them within a given application and across the organization. Sonia starts with some definitions since the term AI is so broad. It's not just about large language models! The conversation also dives into carbon footprints, and using the FinOps models to help with costing for your cloud infrastructure, whether you're using AI or not!
Links
- OpenAI Tokenizer
- Manage Cost for Azure OpenAI Service
- Sustainable Software Engineering with AKS
- Azure Carbon Optimization
- FinOps Foundation
- FinOps FOCUS
- Phoenix Project
Recorded February 19, 2025
Similar Episodes
Apr 10, 2026 ·62m
Apr 9, 2026 ·101m
Apr 8, 2026 ·21m
Apr 7, 2026 ·52m
Apr 3, 2026 ·63m