pablo90salgado

Coordinates multi-agent systems running in parallel or sequence. Defines roles, manages task dependencies, prevents context conflicts and consolidates results from subagents.

Debugs errors and traces failures in AI agents and their tools. Identifies context errors, infinite loops, malformed tool calls, response parsing issues and subagent conflicts.

Designs and runs evaluation pipelines for AI models and agents. Builds benchmarks, computes prompts, calculates accuracy and produces comparison reports between versions.

Analyzes, cleans, and describes CSV files and tabular data, detecting data types, null values, duplicates, and outliers. It also generates executive summaries in plain text or markdown.