Can AI Agents Generate Microservices? How Far are We?

2026-03-09

AI Summary

This study presents the first systematic evaluation of large language model (LLM) agents in automatically generating microservices with explicit dependencies and API contracts. Through 144 experiments, we compare three LLM agents, two prompting strategies, and two generation scenarios—incremental versus from-scratch—using unit and integration tests to assess functional correctness, code quality, and efficiency. Results show that incremental generation achieves unit test pass rates of 50–76%, while clean-state generation yields integration test pass rates of 81–98%. Although the generated code exhibits lower complexity than human-written baselines, its correctness remains inconsistent, necessitating human oversight. The findings further reveal that both prompting strategy and generation scenario significantly influence adherence to API contracts and overall code quality.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: Code Generation / Program Synthesis from Natural Language
• Multiagent Systems: Modeling other Agents

Application Category

• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

LLMs have advanced code generation, but their use for generating microservices with explicit dependencies and API contracts remains understudied. We examine whether AI agents can generate functional microservices and how different forms of contextual information influence their performance. We assess 144 generated microservices across 3 agents, 4 projects, 2 prompting strategies, and 2 scenarios. Incremental generation operates within existing systems and is evaluated with unit tests. Clean state generation starts from requirements alone and is evaluated with integration tests. We analyze functional correctness, code quality, and efficiency. Minimal prompts outperformed detailed ones in incremental generation, with 50-76% unit test pass rates. Clean state generation produced higher integration test pass rates (81-98%), indicating strong API contract adherence. Generated code showed lower complexity than human baselines. Generation times varied widely across agents, averaging 6-16 minutes per service. AI agents can produce microservices with maintainable code, yet inconsistent correctness and reliance on human oversight show that fully autonomous microservice generation is not yet achievable.

Problem

• AI Agents
• Microservices
• Code Generation
• API Contracts
• LLMs

Innovation

• AI agents
• microservice generation
• LLM prompting strategies
• API contract adherence
• incremental vs clean-state generation