# Fantastic Scientific Agents and How to Build Them: AgentBuild for Rietveld Refinement

arXiv: https://arxiv.org/abs/2606.12834 · HTML: https://mlsanity.com/paper/2606.12834/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Placed in Web, search, and research agents, one of 16 directions in LLM agents and tool use.
- Builds on 21 of its 47 references that are in the map.
- First posted on arXiv 11 Jun 2026.

## Placed in

- LLM agents and tool use / [Web, search, and research agents](https://mlsanity.com/agents/web-search-and-research-agents/)

## Builds on (most cited first, 21 of 21)

- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [ReAct: Synergizing Reasoning and Acting in Language Models](https://mlsanity.com/paper/2210.03629/) (2022)
- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [Reflexion: Language Agents with Verbal Reinforcement Learning](https://mlsanity.com/paper/2303.11366/) (2023)
- [Optuna: A Next-generation Hyperparameter Optimization Framework](https://mlsanity.com/paper/1907.10902/) (2019)
- [The AI Scientist: Towards Fully Automated Open-Ended Scientific Discovery](https://mlsanity.com/paper/2408.06292/) (2024)
- [DSPy: Compiling Declarative Language Model Calls into Self-Improving Pipelines](https://mlsanity.com/paper/2310.03714/) (2023)
- [ChemCrow: Augmenting large-language models with chemistry tools](https://mlsanity.com/paper/2304.05376/) (2023)
- [Eureka: Human-Level Reward Design via Coding Large Language Models](https://mlsanity.com/paper/2310.12931/) (2023)
- [GEPA: Reflective Prompt Evolution Can Outperform Reinforcement Learning](https://mlsanity.com/paper/2507.19457/) (2025)
- [Accelerating scientific discovery with Co-Scientist](https://mlsanity.com/paper/2502.18864/) (2025)
- [Parsl: Pervasive Parallel Programming in Python](https://mlsanity.com/paper/1905.02158/) (2019)
- [funcX: A Federated Function Serving Fabric for Science](https://mlsanity.com/paper/2005.04215/) (2020)
- [PROV-AGENT: Unified Provenance for Tracking AI Agent Interactions in Agentic Workflows](https://mlsanity.com/paper/2508.02866/) (2025)
- [Applying the FAIR Principles to computational workflows](https://mlsanity.com/paper/2410.03490/) (2024)
- [LLM Agents for Interactive Workflow Provenance: Reference Architecture and Evaluation Methodology](https://mlsanity.com/paper/2509.13978/) (2025)
- [Do Large Language Models Speak Scientific Workflows?](https://mlsanity.com/paper/2412.10606/) (2024)
- [A Terminology for Scientific Workflow Systems](https://mlsanity.com/paper/2506.07838/) (2025)
- [Towards a Modular Architecture for Science Factories](https://mlsanity.com/paper/2308.09793/) (2023)
- [Employing Artificial Intelligence to Steer Exascale Workflows with Colmena](https://mlsanity.com/paper/2408.14434/) (2024)
- [Dara: Automated multiple-hypothesis phase identification and refinement from powder X-ray diffraction](https://mlsanity.com/paper/2510.19667/) (2025)

## Built on by (most cited first, 3 of 3)

- [ScienceBuddy: Recursive-in-Recursive Self-Improvement for Interactive Scientific Agents](https://mlsanity.com/paper/2609.17523/) (2026)
- [AutoXRD: Autonomous LLM Agents and Comprehensive Evaluation for Powder Diffraction Analysis](https://mlsanity.com/paper/2609.00070/) (2026)
- [Toward Trustworthy Autonomous Science: A Two-Year Community Roadmap](https://mlsanity.com/paper/2607.12113/) (2026)

## Abstract (arXiv)

As scientific workflows shift from deterministic executables to LLM-based agents, the development practices on offer, such as fine-tuning, reinforcement learning, and prompt-and-go, bury the scientist's judgment. We propose treating agent construction as a workflow stage and introduce AgentBuild, which builds a scientific agent from a contract the scientist authors. The contract is a version-controlled rubric, a difficulty-graded curriculum, and a curated external knowledge base. A rubric-driven judge gates a meta-optimizer coding agent that edits the agent within a declared boundary, so the build compiles the agent, not the scientist's judgment. We instantiate this for Rietveld refinement of X-ray diffraction data through GSAS-II behind MCP and A2A, where a blank-harness construction run progresses through a lithium lanthanum zirconium oxide (LLZO) signal-to-noise ladder, reaches the 4 hour scan as a frontier case, and exposes the workflow-scope limits that remain. The same rubric that rewards credible fits also scores trajectory scope, making the frontier a contract failure rather than a pattern-fitting failure. As base models evolve, re-running AgentBuild is a re-tune, not a rebuild, and the scientist's authored contract remains the durable asset.