About the webinar
In this MLOps and LLMOps webinar, we’ll walk through the entire AI lifecycle – from idea and experimentation to production, deployment and continuous monitoring, highlighting how AI differs from traditional software (data-driven, non-linear, and sometimes unpredictable even when “done right”). You’ll learn the main deployment patterns (batch/offline, real-time/online, and common patterns employed in cloud solutions) and the key trade-offs around latency, scaling, and operational reliability.
We’ll then connect MLOps and LLMOps in a practical way: versioning data/models/prompts, reproducibility, CI/CD, and testing strategies for probabilistic systems. It’s aimed at data scientists, ML engineers, software engineers, and AI engineers who want a clear, production-focused view of how to run ML and LLM solutions end-to-end.
Who is the webinar for?
It’s aimed at data scientists, ML engineers, software engineers, and AI engineers who want a clear, production-focused view of how to run ML and LLM solutions end-to-end. Also suited to those with no experience in building and deploying AI models, and are curious on AI/ML/LLM Ops.
Key takeaways for participants:
-
Key differences between AI and traditional software
-
How these differences translate to model deployment
-
What is ML and LLM Ops and how they differ
-
Different model deployment strategies
Speaker bio:
Murilo Kuniyoshi Suzart Cunha (https://www.linkedin.com/in/murilo-cunha/)
Murilo is a machine learning engineer specializing in productionizing models and applying AI Ops best practices, with a focus on the evolving landscape of LLMOps. He takes a pragmatic approach to machine learning, ensuring AI initiatives deliver tangible ROI. An experienced international conference speaker and open source supporter, Murilo is also the host of the Monkey Patching Podcast.