Blog

Evaluating LLM-Based Agents
INTRODUCTION

TaskGenAgent for Agile Task Management

TaskGenAgent is an AI agent for task generation within the HIVEMIND project, breaking down software specifications into prioritized, structured, and actionable development tasks. Developed at UPC with Carles Farré, Lidia López, Xavier Franch, and Razieh Akbari.

Read More
Evaluating LLM-Based Agents
ARTICLE

Generative UI for TaskGenAgent: Designing Interactive Support for Agile Task Management

How LangChain’s Generative UI approach can support TaskGenAgent by turning generated software tasks into structured, interactive, and reviewable Agile task management artefacts.

Read More
Evaluating LLM-Based Agents
ARTICLE

Evaluating LLM-Based Agents

Agents From Observability to Continuous Improvement Hover to Analyze Evaluating LLM-Based Agents LLM-based agents are multi-step and non-deterministic entities. Meaningful evaluation requires understanding why and how an agent arrived at a result. Some tools like Arize Phoenix , provide structured mechanisms to visualize these behaviors.

Read More