# AI for Distributed Systems Design: Scalable Cloud Optimization Through Repeated LLMs Sampling And Simulators

arXiv: https://arxiv.org/abs/2510.18897 · HTML: https://mlsanity.com/paper/2510.18897/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 8 of its 18 references that are in the map.
- First posted on arXiv 20 Oct 2025.

## Builds on (most cited first, 8 of 8)

- [Large Language Monkeys: Scaling Inference Compute with Repeated Sampling](https://mlsanity.com/paper/2407.21787/) (2024)
- [Draft, Sketch, and Prove: Guiding Formal Theorem Provers with Informal Proofs](https://mlsanity.com/paper/2210.12283/) (2022)
- [Barbarians at the Gate: How AI is Upending Systems Research](https://mlsanity.com/paper/2510.06189/) (2025)
- [Performance Prediction for Large Systems via Text-to-Text Regression](https://mlsanity.com/paper/2506.21718/) (2025)
- [Building a serverless Data Lakehouse from spare parts](https://mlsanity.com/paper/2308.05368/) (2023)
- [Safe, Untrusted, "Proof-Carrying" AI Agents: toward the agentic lakehouse](https://mlsanity.com/paper/2510.09567/) (2025)
- [LAVA: Lifetime-Aware VM Allocation with Learned Distributions and Adaptation to Mispredictions](https://mlsanity.com/paper/2412.09840/) (2024)
- [Eudoxia: a FaaS scheduling simulator for the composable lakehouse](https://mlsanity.com/paper/2505.13750/) (2025)

## Built on by (most cited first, 1 of 1)

- [Test-Time Optimization of Physical Query Plans with LLMs](https://mlsanity.com/paper/2602.10387/) (2026)

## Abstract (arXiv)

We explore AI-driven distributed-systems policy design by combining stochastic code generation from large language models (LLMs) with deterministic verification in a domain-specific simulator. Using a Function-as-a-Service runtime (Bauplan) and its open-source simulator (Eudoxia) as a case study, we frame scheduler design as an iterative generate-and-verify loop: an LLM proposes a Python policy, the simulator evaluates it on standardized traces, and structured feedback steers subsequent generations. This setup preserves interpretability while enabling targeted search over a large design space. We detail the system architecture and report preliminary results on throughput improvements across multiple models. Beyond early gains, we discuss the limits of the current setup and outline next steps; in particular, we conjecture that AI will be crucial for scaling this methodology by helping to bootstrap new simulators.