Скачать или смотреть ML Real Talk Episode 2: MLOps Best Practices in the LLM era

It’s not enough for data scientists and ML engineers to build and customize models, companies want to put those models into production and generate real world impact. However, successfully deploying and managing AI/ML models at scale can be a complex challenge. That's where MLOps comes in.

During this interactive discussion with Mikiko Bazeley, head of MLOps at Featureform, we explored what it takes to build successful MLOps practices in the LLM era. Throughout her career, Mikiko has successfully led numerous projects, bridging the gap between data science and engineering teams. Her passion for optimizing AI/ML workflows and driving operational excellence has resulted in streamlined processes, accelerated model deployment, and enhanced collaboration.

During this session, we covered a number of topics including:
MLOps Best Practices: Mikiko's extensive experience allows her to share proven strategies and best practices for implementing successful MLOps methodologies.
Real-World Examples: With a track record of delivering tangible results, Mikiko will draw from her experience in deploying ML models in production, providing practical insights and real-world case studies including recent experience with LLMs.
Collaboration: Mikiko's expertise in fostering collaboration between data science and engineering teams will shed light on the critical importance of effective communication and streamlined workflows.
Cutting-Edge Tools: As an industry professional, Mikiko stays up-to-date with the latest MLOps tools and technologies, offering guidance on selecting the right tools for your organization's specific needs including what does MLOps look like in the LLM era.