mlsanityData to 14 Oct 2026

ML Engineering Agents - A Data Management Perspective

arXiv 2610.16919 ↗First posted 12 Oct 2026cs.DB cs.LG cs.SEcited by 0

On the map: LLM agents and tool use → Software engineering coding agents

Abstract (arXiv)

Machine learning engineering (MLE) agents are emerging as a new paradigm for automating end-to-end ML pipeline development. Using large language models for code generation and predictive performance as a machine-gradable feedback signal, these agents iteratively synthesize, execute, and refine pipeline code. Yet current systems are largely evaluated on pre-built, single-table benchmarks that abstract away key data management challenges. This tutorial introduces the foundations of MLE agents and highlights research opportunities for the database community. We cover how agents combine program synthesis with iterative and evolutionary optimization, and discuss open challenges in efficient pipeline execution, data discovery and augmentation over heterogeneous sources, reliability and debugging, as well as fairness and compliance. We connect these challenges to existing data management techniques and conclude with a hands-on session using contemporary MLE agents.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Software engineering coding agents, one of 16 directions in LLM agents and tool use. Builds on 34 of its 56 references that are in the map. First posted on arXiv 12 Oct 2026.

Builds on

34 of 56 references.

Closest ancestors · 2

Same direction, newest first

  1. 2025AccelOpt: A Self-Improving LLM Agentic System for AI Accelerator Kernel Optimization19 Nov
  2. 2025AlphaEvolve: A coding agent for scientific and algorithmic discovery16 Jun

Other references · 32

  1. 2023Large Language Models as Optimizerscited by 864
  2. 2024MLE-bench: Evaluating Machine Learning Agents on Machine Learning Engineeringcited by 469
  3. 2015Illuminating search spaces by mapping elitescited by 425
  4. 2025AIDE: AI-Driven Exploration in the Space of Codecited by 226
  5. 2023Large Language Models for Automated Data Science: Introducing CAAFE for Context-Aware Automated Feature Engineeringcited by 127
  6. 2024Next-Generation Database Interfaces: A Survey of LLM-based Text-to-SQLcited by 120
  7. 2024Why Tabular Foundation Models Should Be a Research Prioritycited by 106
  8. 2025AI Research Agents for Machine Learning: Search, Exploration, and Generalization in MLE-benchcited by 71
  9. 2026Towards a Science of AI Agent Reliabilitycited by 65
  10. 2025MLE-STAR: Machine Learning Engineering Agent via Search and Targeted Refinementcited by 58
  11. 2024BEAVER: An Enterprise Benchmark for Text-to-SQLcited by 42
  12. 2025Supporting Our AI Overlords: Redesigning Data Systems to be Agent-Firstcited by 32
20 more
  1. 2026MLEvolve: A Self-Evolving Framework for Automated Machine Learning Algorithm Discoverycited by 28
  2. 2025MLZero: A Multi-Agent System for End-to-end Machine Learning Automationcited by 24
  3. 2021Production Machine Learning Pipelines: Empirical Analysis and Optimization Opportunitiescited by 23
  4. 2019SystemDS: A Declarative Machine Learning System for the End-to-End Data Science Lifecyclecited by 20
  5. 2022Data Debugging with Shapley Importance over End-to-End Machine Learning Pipelinescited by 17
  6. 2023DiffPrep: Differentiable Data Preprocessing Pipeline Search for Learning over Tabular Datacited by 9
  7. 2022DiffML: End-to-end Differentiable ML Pipelinescited by 7
  8. 2025Deep Research is the New Analytics System: Towards Building the Runtime for AI-Driven Analyticscited by 7
  9. 2026Bespoke OLAP: Synthesizing Workload-Specific One-size-fits-one Database Enginescited by 7
  10. 2024Retrieve, Merge, Predict: Augmenting Tables with Data Lakescited by 5
  11. 2025Beyond Relational: Semantic-Aware Multi-Modal Analytics with LLM-Native Query Optimizationcited by 5
  12. 2026Data Agents: Levels, State of the Art, and Open Problemscited by 4
  13. 2023Kitana: Efficient Data Augmentation Search for AutoMLcited by 3
  14. 2026SemPipes -- Optimizable Semantic Data Operators for Tabular Machine Learning Pipelinescited by 3
  15. 2026Recovering Wasted Compute in Autoresearch Agentscited by 2
  16. 2025Rethinking Query Optimization for Multi-Agent Systems [Vision]cited by 1
  17. 2026stratum: A System Infrastructure for Massive Agent-Centric ML Workloadscited by 1
  18. 2026Be Fair! Can Machine Learning Engineering Agents Adhere to Fairness Constraints?cited by 1
  19. 2026Demonstrating GenDB: Instance-Optimized and Customized Query Processing Code Generation via LLM Agentscited by 1
  20. 2026Walk Before You Run: The Importance of Data Exploration for Data Analysis Agentscited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025R&D-Agent: An LLM-Agent Framework Towards Autonomous Data Science
  2. 2025MLE-Smith: Scaling MLE Tasks with Automated Multi-Agent Pipeline
  3. 2025ML-Tool-Bench: Tool-Augmented Planning for ML Tasks
  4. 2026DataClawEval: A Benchmark for Data Engineering Agents in Real Industrial Harness
  5. 2025ArchPilot: A Proxy-Guided Multi-Agent Approach for Machine Learning Engineering
  6. 2026kRAIG: A Natural Language-Driven Agent for Automated DataOps Pipeline Generation
  7. 2025MLE-Dojo: Interactive Environments for Empowering LLM Agents in Machine Learning Engineering
  8. 2026Data Intelligence Agents: Interpreting, Modeling, and Querying Enterprise Data via Autonomous Coding Agents