Declarative Data Services: Structured Agentic Discovery for Composing Data Systems

2026-05-20

AI Summary

This work addresses the challenge of stabilizing convergence to feasible data system configurations in heterogeneous multi-system backends, where conventional unconstrained agents often fail. The authors propose a structured agent discovery framework that decomposes global search into bounded subproblems through a four-layer typed contract architecture, enabling forward knowledge propagation and backward error attribution while supporting inline skill referencing and runtime attribution. Integrating large language models, declarative intent parsing, operator DAG generation, and contract-driven error repair, the framework significantly enhances convergence on transactional backend workloads. It not only achieves stable deployment but also transforms runtime failures into reusable skill patches that can be inlined in subsequent tasks.

Abstract

Agentic discovery has shown that LLM-driven search can find novel algorithms, designs, and code under benchmark conditions. Translating the paradigm to multi-system data backends surfaces a harder problem: the search space is heterogeneous, the verifier is whether a deployed stack actually runs, and composition knowledge is unevenly captured in pretraining. Unbounded agentic discovery, a coding agent iterating on failure-log feedback, fails to converge consistently on a working stack even when iteration and explicit composition knowledge are added. We propose Declarative Data Services (DDS), an architecture for structured agentic discovery of data-system compositions from declarative user intent. The framework owns four typed contracts at successive layers (intent, operator DAG, per-system skills, runtime attribution) that decompose the global search into bounded sub-searches; sub-agents search each typed space, while the framework provides the channels by which knowledge flows forward as inline skill citations and errors route backward as typed signals. As a proof of life on a trading-backend workload, DDS converges where unbounded discovery does not; runtime failures become skill patches that the next deployment cites inline. We position this as an early prototype reporting lessons from real-world data-system composition.

Problem

• agentic discovery
• data systems composition
• declarative intent
• heterogeneous search space
• runtime verification

Innovation

• Declarative Data Services
• Agentic Discovery
• Structured Search
• Data System Composition
• Typed Contracts