Главная
Study mode:
on
1
[] Rohit's preferred coffee
2
[] Takeaways
3
[] Please like, share, and subscribe to our MLOps channels!
4
[] Rohit's current work
5
[] The Portkey landscape
6
[] Compute unit is no longer a Cloud resource, it's a Foundational Model
7
[] Hang-ups at high-scale models and how to combat them
8
[] Complexity of the Apps evolving
9
[] Rohit's working relationships with the agents
10
[] Fine-tuning reliability
11
[] Small language models can outperform larger ones
12
[] Market map at Portkey
13
[] AI Gateway
14
[] Worker Bee and Queen Bee
15
[] Security and Compliance
16
[] Idea of Data Mesh
17
[] Forward compatibility
18
[] Decoupling AI Gateway from the code
19
[] Hardest design decisions to make since creating Portkey
20
[] Wrap up
Description:
Save Big on Coursera Plus. 7,000+ courses at $160 off. Limited Time Only! Grab it Explore the intricacies of designing for forward compatibility in generative AI with Rohit Agarwal, CEO of Portkey.ai, in this insightful MLOps podcast episode. Delve into practical strategies for securing LLM systems and gaining confidence in production deployments. Learn about the evolving landscape of AI infrastructure, including the shift from cloud resources to foundational models as compute units. Discover solutions for high-scale model challenges, the increasing complexity of AI applications, and the potential of small language models. Gain valuable insights on AI gateways, security and compliance measures, and the concept of data mesh. Understand the importance of forward compatibility and the benefits of decoupling AI gateways from code. Rohit shares his experiences and hard-learned lessons in creating Portkey, offering a comprehensive look at the current state and future direction of MLOps and generative AI deployment.

Designing for Forward Compatibility in Gen AI - MLOps Podcast

MLOps.community
Add to list