Ph.D. Candidate & Applied AI Engineer
Zakaria Hachm
IMT Atlantique • LS2N (UMR CNRS 6004)
PhD candidate designing reliable, tool-using LLM agents for engineering software.
Agent evaluation/benchmarkingModel TransformationsAgentic AISelf-improving agentsSynthetic data generation
Publications1
Projects5
Courses3
Featured Projects
All ProjectsEMF Server
A stateless API server exposing a set of routes for metamodel-independent operations. It enables generic and dynamic path operations on any model element, providing interoperability between web applications and EMF models.
Model ManagementEMFREST API
Model Management Agents
LLM-based agents for model transformation (ATL) and model editing (EMF), exposed via MCP servers. Combines dedicated per-tool interfaces with a relevance-score-based tool filtering mechanism to improve tool selection accuracy and scalability as the number of available tools grows. Includes a reusable dataset generation methodology and 1,200+ labeled instruction–tool-call pairs for evaluation.
PythonLangGraphATLEMFMCPLLM Agents
Megamodel-based Agent Evaluation
A megamodel-based approach for evaluating ecosystems of LLM-based modeling agents. Unifies modeling artifacts, tools, agent workflows, and execution traces in a single structured repository, supporting three complementary capabilities: automated model discovery from existing execution logs, dataset augmentation from small seed samples, and systematic agent benchmarking. Validated on LLM-based agents for ATL transformation and EMF model-handling tasks, discovering 16,000+ megamodel elements from real execution logs and enabling regression testing across agent versions.
PythonLangSmithATLEMFLLM Agents
Featured Publications
All PublicationsSAM 2025 • 2025
Towards LLM Agents for Model-Based Engineering: A Case in Transformation Selection
Zakaria Hachm, Théo Le Calvar, Hugo Bruneliere, Massimo Tisi
In Model-Based Engineering (MBE), practitioners frequently face the challenge of selecting appropriate tools from a large number of options. This requires both deep domain-specific knowledge and technical expertise. LLM-based agents are software components that depend on Large Language Models (LLMs) to autonomously select and apply software tools to perform specific tasks.