AI Explained: When Frontier Models Are Overkill
November 17, 2026
10:00AM PT / 1:00PM ET
Registration is now closed. Please check back later for the recording.
Many teams build agents and guardrails on the largest model available because bigger is seen as better. That holds for complex reasoning, but not for focused tasks like guardrails, where a smaller model built for the job can perform just as well at a fraction of the cost and latency. As more agents go live, choosing the right model size affects how reliably they run, not just what they cost. Join Javier Fernandez Rico, Head of Engineering at Fastino, for a conversation about when frontier models are the right call, when smaller models do the work better, and what that means for keeping agents safe in production.
What you’ll learn:
- How to tell which parts of an agentic system need a frontier model and which run better on a task-specific one
- What to look for when evaluating a guardrail model, from accuracy and speed to where it can be deployed
- Why model pricing shapes architecture decisions more than most teams expect, and how leaders can plan around it
AI Explained is our AMA series featuring experts on the most pressing issues facing Agentic and AI teams.
Featured Speakers
Javier Fernandez Rico
Head of Engineering
at
Fastino Labs
Javier Fernandez Rico is Head of Engineering at Fastino. He has spent his career building AI systems that run in the real world, from robotics to interactive entertainment, with more than a dozen patents and research published at ICRA.
Joshua Rubin
Head of AI Science
at
Fiddler AI
Joshua Rubin is Head of AI Science at Fiddler AI, an enterprise AI Observability company. He’s built and led a data science team that developed novel explainability tools for computer vision and multimodal deep-learning models, and techniques for measuring model robustness and drift in unstructured data, key components of Fiddler's LLM observability product. Most recently he's been developing small BERT-scale models to close the feedback loop on measuring large language model performance, serving customers including cloud-native travel platforms, large financial services firms, ad-tech companies, and cryptocurrency exchanges.