Optimizing agentic AI workflows: Metrics-driven evaluation with W&B Weave and NVIDIA - FC London 25 @WeightsBiases
Optimizing agentic AI workflows: Metrics-driven evaluation with W&B Weave and NVIDIA - FC London 25  @WeightsBiases
Uploaded December 2025 | Updated September 2026, 2 weeks ago
In this session from Fully Connected London, Rita Fernandes Neves, Sr. Solutions Architect at NVIDIA, explores how to build, debug, and optimize agentic workflows—where large language models drive multi-step reasoning and tool use. She demonstrates how W&B Weave and the NVIDIA NeMo Agent Toolkit enable transparent, metrics-focused development of agentic applications.

Rita showcases practical strategies for workflow instrumentation, evaluation, and visualization across structured operations and open-ended research assistants. Learn how YAML-driven configuration, automated and human-in-the-loop evaluation, and intuitive dash boarding help teams benchmark models, diagnose bottlenecks, and iteratively refine AI agents for reliability and performance.
Optimizing agentic AI workflows: Metrics-driven evaluation with W&B Weave and NVIDIA - FC London 25Fully Connected 2025 kickoff: The rise (and the challenges) of the agentic eraIs there a Better Way to Code?Let Claude Code add observability to your agentLocal or Global? The Big Question in AI Drug Discovery | Isomorphic LabsPassenger or Pedestrian: Who Will an AI Car Save?Why Anthropic, Meta, and Tesla All Chose the Same Database | Aaron Katz, ClickHouseW&B Models: Media Tracking and Sync📅 ThursdAI - Claude Code leak w Sigrid, Veo 3.1 Lite & more AI newsShe Raised $64M to Build an AI Math Prodigy | Carina Hong, CEO of AxiomProduction monitoring for AI applications using W&B Weave📅 ThursdAI - LIVE from AI Engineer Worlds Fair - OpenAI, DeepMind, EXO, Sakana & more friends
Weights & Biases |

Optimizing agentic AI workflows: Metrics-driven evaluation with W&B Weave and NVIDIA - FC London '25

SHARE TO X SHARE TO REDDIT SHARE TO FACEBOOK WALLPAPER