Scientific Python, MLIP & Atomistic Simulation Workflow Development

Scientific software development and computational workflow engineering for materials researchers and industrial R&D teams. The focus is on improving simulation automation, numerical reliability, verification, and reproducibility across atomistic modelling and machine learning workflows. Technical Capabilities:

• Python-based scientific software and ASE integrations.
• Quantum ESPRESSO workflow automation and data processing.
• MLIP training, evaluation, and benchmarking workflows.
• NEB/CI-NEB reaction-path and diffusion-barrier analysis.
• Active learning, uncertainty quantification, and scientific data pipelines.
• Automated testing, numerical validation, and provenance tracking. Existing Research Tools: My development work includes ActiStruct, NEBWalk, and MLIPForge, covering computational research orchestration, transition-state analysis, and MLIP-related workflows. These are research-stage software efforts, with deployment readiness evaluated according to project requirements. Collaboration Opportunities: Available for custom research software development, workflow integration, computational benchmarking, reproducibility audits, and collaborative R&D. Deliverables are defined by scope and may include documented Python software, validated test cases, scientific reports, and reproducible workflows.