A system that succeeds once is a demo. A system that succeeds every time is a breakthrough. Dr. Danielle Perszyk sits down with AI researcher Deniz Birlikci from Amazon's AGI Lab to explore how reinforcement learning (RL) is transforming AI agents from impressive demos into dependable tools that work consistently in real-world environments.

Danielle and Deniz discuss why reliability, not accuracy, is the true bottleneck for web agents, the critical role of a robust verification system, failure models that RL attempts to fix, and the extraordinary complexity of orchestrating live browsers with perception and actuation stacks. Discover how RL is building the foundation for agents that can handle complex workflows reliably alongside humans.

Before an AI agent can reason or plan, it has to see. Dr. Danielle Perszyk and AI researcher Matthew Elkherj explore why user interface (UI) understanding is one of the most underestimated challenges in building autonomous agents—and why it’s foundational to creating reliable AI teammates.

Danielle and Matthew discuss the distinct reliability requirements of agents, how perceptual hallucinations can be a feature (rather than a bug), and the role of synthetic gym environments in training. Together, they explain why building reliable agents requires solving interconnected challenges—from how agents perceive digital interfaces to how they learn from mistakes, handle real-world complexity, and ultimately augment human capacity.

Please note: this podcast was recorded in August 2025.

How do you build environments complex enough to train agents that can handle the real web? Dr. Danielle Perszyk sits down with Jason Laster, an engineer leading Amazon's AGI Lab's effort to build reinforcement learning (RL) gyms— simulated web environments where agents learn—to explore how environment development is as critical as models, data, and compute. The browser is one of the most complex worlds we could possibly train in, and this conversation unpacks why high-fidelity simulations that capture every UI quirk matter more than building thousands of basic environments. Discover how RL gyms are finally becoming practical at scale, why observability and verifiable rewards are essential for rigorous training, and why simulated environments beat the real web for developing reliable autonomous systems.

How did we get from language models to AI agents that can take action? Dr. Danielle Perszyk sits down with Kelsey Szot, product lead at Amazon's AGI Lab and one of the original founders of Adept (a startup that helped pioneer modern AI agents) to discuss the technical breakthroughs that transformed AI from pattern recognition to agentic capabilities. Danielle and Kelsey trace the evolution from early distributed training at scale to today's autonomous systems that can reason, plan, and interact with real environments—exploring the shift from rigid, rules-based automation to AI that can generalize across changing interfaces and complex workflows.

Coming soon!