Driving · 2026-10-06

MiniCorp: The Last Mile of the AI Agent Firm

Jingying Zeng, Zhenwei Dai, Jinning Li, Changho Shin, Dylan Zhang, Yuxuan Lu, Qi He, Dakuo Wang, Kai-Wei Chang

arXiv:2610.05912PDF

AI-assisted summary based on the paper. Consult the linked source for methods and limitations. Benchmark numbers appear after editor verification.

TL;DR

MiniCorp is a closed-loop office simulator where a persistent multi-role agent firm runs an e-commerce company against an evolving external market model. It records communications and decisions continuously, preserves information available at the time, and supports checkpointing to replay the same state under alternative decisions. The authors evaluate end-to-end external-world fidelity against empirical market-response patterns and show coordinated role behavior, adaptive learning from market feedback, and sustained advertising exploration with long-term strategic guidance.

Why it matters

The authors target the lack of longitudinal, counterfactual enterprise data for training/evaluating long-horizon organizational agents. MiniCorp couples an agent firm with an external market that continues to evolve independent of firm actions, producing sequential decision–consequence data and enabling replay under alternative decisions via checkpointing. The external world’s fidelity is evaluated system-level against empirical response patterns to reduce simulator overfitting risk.

Method

  • MiniCorp couples two stateful systems: an evolving external market (consumers, competitors, suppliers, platform, macro shocks) and a persistent internal agent firm with organizational roles and decision cycles.
  • In an e-commerce setup, the external world simulates a response chain (search demand, query–listing matching, organic ranking, sponsored auctions, click/conversion, competitor responses) so actions affect downstream intermediate pathways, not only terminal outcomes.
  • The authors evaluate end-to-end external-world fidelity against empirical response patterns and use checkpointing to replay identical situations under different decisions for counterfactual comparisons.
Abstract (from arXiv)

The last mile toward enterprise AGI is a company that runs itself. Training and adapting such agents require longitudinal enterprise data, which remain scarce, costly to acquire, and often restricted by privacy constraints. Historical archives are also frequently incomplete and record only what actually happened. They cannot show the outcomes of alternative decisions. We introduce MiniCorp, an office simulator for studying how agents can collectively run a company while generating enterprise data at scale. Using an e-commerce company as a demonstration, MiniCorp connects two interacting worlds. The external world models customers, dynamic competitors, and market mechanisms. The internal world consists of agents that observe events, discuss their options, and make strategic decisions. These decisions have lasting effects on the market, and the resulting feedback informs the firm's later decisions. As the firm and market interact, MiniCorp continuously records the agents' communications and decisions. These records preserve the information available at the time and the business results that followed. Checkpointing allows the same situation to be replayed under different decisions, providing comparisons unavailable in static archives. We evaluate end-to-end fidelity against patterns reported in empirical studies of real markets. These evaluations provide agents with realistic market feedback and reduce the risk that they learn to exploit flaws in the simulator. Our experiments show agents coordinating across roles and adapting their decisions to market feedback. With explicit long-term strategic guidance, they also sustain advertising exploration despite weak early returns. MiniCorp thus provides an environment for studying AI-run companies and a scalable source of longitudinal and counterfactual enterprise data for agent training and evaluation.

Related papers

Publication record

Bibliographic evidence describes where and when a record appeared. It does not validate scientific claims.

First public release: 2026-10-05 · Indexed by Pub-AI: 2026-10-06 UTC

preprint

MiniCorp: The Last Mile to an AI Agent Corporation

Jingying Zeng, Zhenwei Dai, Jinning Li, Changho Shin, Dylan Zhang, Yuxuan Lu, Qi He, Dakuo Wang, Kai-Wei Chang

Preprint posted: 2026-10-05 · Original record ↗

arxiv
2610.05912
Explicit source versions (1)
  • v2 · 2026-10-13
  1. Explicit source version detected · v2

    Source date: 2026-10-13 · First detected: 2026-10-25 UTC · Historical baseline

    Source evidence (1)
  2. Preprint available

    Source date: 2026-10-05 · First detected: 2026-10-25 UTC · Historical baseline

    Source evidence (1)
Abstract · arxiv

The last mile toward enterprise AGI is a company that runs itself. Training and adapting such agents require longitudinal enterprise data, which remain scarce, costly to acquire, and often restricted by privacy constraints. Historical archives are also frequently incomplete and record only what actually happened. They cannot show the outcomes of alternative decisions. We introduce MiniCorp, an office simulator for studying how agents can collectively run a company while generating enterprise data at scale. Using an e-commerce company as a demonstration, MiniCorp connects two interacting worlds. The external world models customers, dynamic competitors, and market mechanisms. The internal world consists of agents that observe events, discuss their options, and make strategic decisions. These decisions have lasting effects on the market, and the resulting feedback informs the firm's later decisions. As the firm and market interact, MiniCorp continuously records the agents' communications and decisions. These records preserve the information available at the time and the business results that followed. Checkpointing allows the same situation to be replayed under different decisions, providing comparisons unavailable in static archives. We evaluate end-to-end fidelity against patterns reported in empirical studies of real markets. These evaluations provide agents with realistic market feedback and reduce the risk that they learn to exploit flaws in the simulator. Our experiments show agents coordinating across roles and adapting their decisions to market feedback. With explicit long-term strategic guidance, they also sustain advertising exploration despite weak early returns. MiniCorp thus provides an environment for studying AI-run companies and a scalable source of longitudinal and counterfactual enterprise data for agent training and evaluation.