Adaptive Data Flywheel: Applying MAPE Control Loops to AI Agent Improvement

TL;DR

This paper presents a practical implementation of an adaptive data flywheel using MAPE control loops to improve enterprise AI agents, demonstrating continuous learning, failure mitigation, and significant latency and accuracy gains in a real-world setting.

Contribution

It introduces a structured MAPE-driven data flywheel framework for enterprise AI, enabling continuous, automated improvements based on real-world feedback and failure analysis.

Findings

Achieved 96% accuracy in routing with a fine-tuned 8B model

Reduced latency by up to 70% through targeted fine-tuning

Identified and addressed key failure modes in RAG pipelines

Abstract

Enterprise AI agents must continuously adapt to maintain accuracy, reduce latency, and remain aligned with user needs. We present a practical implementation of a data flywheel in NVInfo AI, NVIDIA's Mixture-of-Experts (MoE) Knowledge Assistant serving over 30,000 employees. By operationalizing a MAPE-driven data flywheel, we built a closed-loop system that systematically addresses failures in retrieval-augmented generation (RAG) pipelines and enables continuous learning. Over a 3-month post-deployment period, we monitored feedback and collected 495 negative samples. Analysis revealed two major failure modes: routing errors (5.25\%) and query rephrasal errors (3.2\%). Using NVIDIA NeMo microservices, we implemented targeted improvements through fine-tuning. For routing, we replaced a Llama 3.1 70B model with a fine-tuned 8B variant, achieving 96\% accuracy, a 10x reduction in model size, and 70\% latency improvement. For query rephrasal, fine-tuning yielded a 3.7\% gain in accuracy and a 40\% latency reduction. Our approach demonstrates how human-in-the-loop (HITL) feedback, when structured within a data flywheel, transforms enterprise AI agents into self-improving systems. Key learnings include approaches to ensure agent robustness despite limited user feedback, navigating privacy constraints, and executing staged rollouts in production. This work offers a repeatable blueprint for building robust, adaptive enterprise AI agents capable of learning from real-world usage at scale.