AT&T faced a significant challenge with 8 billion daily tokens, prompting a shift in their AI orchestration approach. Chief Data Officer Andy Markus led the adoption of a multi-agent stack on LangChain, revolutionizing the orchestration layer. Large language model ‘super agents’ now direct smaller, purpose-driven ‘worker’ agents, boosting efficiency and reducing costs by up to 90%, as reported by VentureBeat.
Markus highlighted the success of Ask AT&T Workflows, a drag-and-drop agent builder leveraging Microsoft Azure. By utilizing proprietary tools for document processing, natural language-to-SQL conversion, and image analysis, AT&T empowers employees to automate tasks with data-driven decisions.
Embracing agile coding methods, AT&T pioneers ‘AI-fueled coding,’ streamlining software development. This approach, akin to RAG, accelerates coding processes and enhances production-grade output. Markus envisions a future where AI-driven coding transforms software development cycles, enabling rapid prototyping and improving productivity across technical and non-technical teams.
Source: VentureBeat