Part-Time AI Benchmark Tasks Architect for Terminal Science

OpenTrain AI, Inc.; United States; Remote; USD 55,000 - 110,000; Part time

Job summary

OpenTrain AI, Inc. seeks a part-time contractor to create realistic terminal-based scientific tasks for AI training and evaluation. You will build reproducible benchmark environments and assess agents’ ability to reason, use tools, and debug calculations.

The role requires strong scientific programming expertise and independent validation of workflows, running in Linux/terminal environments with fixed dependencies.

Qualifications

• PhD or equivalent advanced technical experience in physical sciences.
• Ability to independently create and validate computational scientific workflows.
• Strong programming skills in Python, C/C++, Julia, Bash, or similar.

Responsibilities

• Design multi-step terminal tasks based on real physical-science workflows.
• Build self-contained computational environments with fixed dependencies and scientific software.
• Create datasets, molecular structures, simulation settings, and model configurations.
• Write expert solutions using Python, Bash, C/C++, Julia, or domain-specific tools.
• Develop automated tests for numerical accuracy, physical consistency, convergence, and output structure.
• Create tasks involving simulations, numerical modeling, data fitting, optimization, spectroscopy, molecular analysis, and visualization.
• Define numerical tolerances, units, boundary conditions, and expected scientific behavior.
• Validate reproducibility and debug dependency, precision, solver stability, performance, and file-format problems.

Skills

Python

C/C++

Julia

Bash

Command-line

Education

PhD or equivalent advanced technical experience

Postdoctoral experience

Tools

Docker

Conda

Git

CI

HPC environments

Job description

OpenTrain AI, Inc. seeks a part-time contractor to create realistic terminal-based scientific tasks for AI training and evaluation. You will build reproducible benchmark environments and assess agents’ ability to reason, use tools, and debug calculations.

The role requires strong scientific programming expertise and independent validation of workflows, running in Linux/terminal environments with fixed dependencies.