Safety
Lyft built 8 agents that resolve 35% of customer issues end-to-end. That stat sounds crazy, but it's the kind of numbers you see when teams …
Lyft built 8 agents that resolve 35% of customer issues end-to-end. That stat sounds crazy, but it's the kind of numbers you see when teams actually close the evals feedback loop. Looking forward to I
Lyft built 8 agents that resolve 35% of customer issues end-to-end. That stat sounds crazy, but it's the kind of numbers you see when teams actually close the evals feedback loop. Looking forward to Interrupt where we'll learn more about how they did it 🚗 @Lyft is heading to Interrupt. Over the last year, Lyft built 8 AI agents that are capable of fully resolving 35% of all customer issues. Head of Data Science, Safety and Customer Care Nick Ung will take us through the evals used internally, how they scale evals with LangSmith…
Related
- Each workflow in Thoth is a full LangGraph agent which can call subagents which themselves are LangGraph agents. Just describe what you want…
- Dejavu: Towards Experience Feedback Learning for Embodied Intelligence
- No More Stale Feedback: Co-Evolving Critics for Open-World Agent Learning
- Beyond Static Sandboxing: Learned Capability Governance for Autonomous AI Agents
- SLALOM: Simulation Lifecycle Analysis via Longitudinal Observation Metrics for Social Simulation
Source: Harrison Chase (X) | 2026-04-21