OpenAI’s Bonnie Xu discusses Kepler, an internal AI data analyst agent built to query 600+ petabytes of data. She explains how they overcome context window limits using MCP, automated code crawling, and RAG. Xu also shares how their team leverages scoped semantic memory for self-learning and utilizes AST-based LLM grading to build a robust, regression-free evaluation pipeline. By Bonnie Xu
The accelerating capabilities of large language models are reaching a point where internal enterprise data analysis can be significantly augmented by sophisticated AI agents.
This move by OpenAI demonstrates a leading-edge application of AI agents for internal data analysis, validating their potential for complex enterprise workflows and setting a precedent for industry adoption.
The paradigm for enterprise data analysis will shift from human-intensive processes to AI-driven workflows, potentially accelerating insights and reducing reliance on traditional data analyst roles.
- · OpenAI
- · Enterprise AI software providers
- · Companies with vast internal data sets
- · Traditional business intelligence platforms
- · Repetitive data analysis services
OpenAI enhances its own operational efficiency and data-driven decision making through Kepler.
Other large enterprises will accelerate their investment in building or acquiring similar AI agent-based data analysis capabilities.
The role of human data analysts will evolve towards overseeing AI agents, managing exceptions, and focusing on higher-level strategic interpretation.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at InfoQ