October 8, 2026
A live conversation with Ahmed Mrad, AI FinOps Lead at a Leading Gaming Company, and Chintu Parikh, VP of Marketing at nOps
AI costs can look manageable during experimentation. But once LLM-powered applications move into production and usage starts to scale, the economics can change quickly.
One of the biggest opportunities is also one of the easiest to overlook: not every request needs your most powerful—and most expensive—model.
Join Ahmed Mrad, AI FinOps Lead at a Leading Gaming Company, and Chintu Parikh, VP of Marketing at nOps, for a practical conversation about optimizing AI costs in production and making smarter decisions around model selection, routing, cost, quality, and observability.
What We’ll Discuss:
-Where teams are overspending as LLM usage moves into production
-How to determine which tasks actually require a frontier model
-When routing requests to lower-cost models makes sense—and when it doesn’t
-Common failure modes when implementing multi-model routing
-How teams can build trust in lower-cost models
-What good AI cost observability looks like across multiple models
-The cost challenges that emerge as AI usage scales
-What teams may be overlooking when it comes to AI cost optimization
-Practical recommendations based on lessons learned so far