No Single Model to Rule Them All: Building Resilient AI Agents Across Open & Closed LLMs

​We’re excited to welcome Emmanuel Acheampong of Crusoe AI to lead a session on building resilient AI agents across both open and closed LLMs.

​Talk Overview:

​AI agents are only as reliable as the models behind them. Most teams start by wiring an agent to a single LLM and calling it done. Then reality hits: rate limits, outages, cost spikes, and tasks where one model underperforms another. The teams building resilient agents in production aren't betting on one model. They're building across many.

This talk covers how to architect AI agents that route intelligently across open and closed LLMs. I'll walk through practical patterns for model selection at inference time: when to use a large frontier model versus a fine-tuned open-weight model, how to build fallback chains that maintain agent quality during provider outages, and how to use routing logic to optimize for cost, latency, and task-specific accuracy. 

​

​Speaker Overview:

​Emmanuel operates at the intersection of frontier AI research, production systems, and global-scale infrastructure. With a focus on translating cutting-edge research into deployed, revenue-generating products, they specialize in scaling AI from experimentation to real-world impact.

​As a founding member of the DevRel team for Managed AI at Crusoe, he works closely with developers to enable the use of high-performance infrastructure for next-generation AI systems. He is also the Co-Founder and former Head of AI at yShade.ai (Google for Startups Black Founders Accelerator ’24), where he led the development of inclusive computer vision systems trained on a 12M+ image dataset to address bias in skin tone detection across industries.

​His work spans building high-performing AI teams, architecting production-grade ML systems, and driving the “last mile” of AI deployment.