This role is for one of our clients

Role Details

• Employment Type: Contract
• Location: Bangladesh, India, Pakistan, Indonesia, Egypt, Ghana, Nigeria, Turkey, Brazil, and Colombia
• Contract Duration: 5 weeks
• Start Date: Immediate
• Work Commitment: 40 hours per week, with at least 4 hours of daily overlap with Pacific Standard Time (PST)
• Compensation: $300 per approved task
• Average Task Completion Time: 6–8 hours
• Contract Benefits: No medical benefits or paid leave

About the Role

We are seeking experienced Scientific Computing and Research Engineering Experts specializing in Earth Sciences to develop realistic, terminal-based scientific computing tasks for AI benchmarking and evaluation.

In this role, you will translate authentic scientific research workflows into challenging, self-contained computational tasks. These tasks will evaluate AI agents' ability to inspect scientific datasets, process geospatial and time-series information, execute numerical models, troubleshoot computational pipelines, and generate scientifically accurate, objectively verifiable outputs.

The ideal candidate combines advanced academic or research expertise in Earth Sciences with strong programming skills and hands-on experience in scientific computing, data processing, and computational modeling. You should be comfortable working in Linux-based terminal environments and building reproducible workflows that reflect real-world scientific challenges.

Requirements

Key Responsibilities

• Scientific Task Development: Convert real-world Earth-science research workflows into well-defined, self-contained terminal-based benchmark tasks.
• Dataset Preparation: Prepare and process climate, atmospheric, geological, hydrological, oceanographic, remote-sensing, and other environmental datasets.
• Computational Environment Setup: Build reproducible environments using appropriate scientific libraries, command-line utilities, and domain-specific tools.
• Reference Solution Development: Create reliable expert solutions using Python, R, Bash, Julia, or other relevant scientific programming languages.
• Scientific Modeling and Analysis: Develop tasks involving geospatial analysis, numerical simulation, time-series processing, interpolation, forecasting, remote sensing, and environmental risk assessment.
• Evaluation and Grading: Define objective evaluation criteria for scientific outputs, numerical accuracy, data transformations, model behavior, and spatial or temporal correctness.
• Scientific Validation: Verify coordinate reference systems, measurement units, timestamps, missing-data handling, scientific assumptions, and numerical consistency.
• Automated Testing: Develop tests to validate numerical tolerances, output formats, metadata, reproducibility, and adherence to task requirements.
• Troubleshooting and Optimization: Diagnose and resolve issues involving geospatial projections, large datasets, dependencies, computational performance, and numerical stability.
• Technical Documentation: Document data provenance, task instructions, expected outputs, edge cases, scientific limitations, and validation procedures.

Required Qualifications

• PhD in an Earth Sciences discipline is mandatory, including climate science, atmospheric science, geophysics, oceanography, geology, hydrology, remote sensing, environmental modeling, or Earth-system science.
• Strong expertise in at least one Earth-science specialization, supported by academic research or relevant professional experience.
• Proficiency in Python, C/C++, MATLAB/Octave, Julia, or similar programming languages used for scientific simulation, modeling, and data analysis.
• Hands-on experience with scientific data processing, numerical modeling, geospatial analysis, environmental datasets, or time-series analysis.
• Experience developing, debugging, and validating computational workflows.
• Comfort working independently in Linux and terminal-based environments.
• Strong understanding of scientific quality control, spatial and temporal data, uncertainty, and numerical accuracy.
• Excellent analytical, problem-solving, and technical documentation skills.

Preferred Qualifications

• Experience with scientific and geospatial libraries such as NumPy, pandas, SciPy, xarray, rasterio, GeoPandas, Cartopy, or GDAL.
• Familiarity with scientific data formats, including NetCDF, HDF5, GeoTIFF, shapefiles, and GRIB.
• Experience working with climate, weather, satellite imagery, seismic, oceanographic, geological, or hydrological datasets.
• Familiarity with Docker, Conda, Git, CI/CD pipelines, automated testing, or high-performance computing (HPC) environments.
• Experience with research software engineering, scientific benchmarking, or automated grading systems.
• Exposure to AI coding agents, terminal-based AI systems, or the development of evaluation tasks for AI models.
• Research publications or open-source contributions in Earth Sciences, environmental science, geospatial analysis, or computational science.

Engagement and Compensation

This is a short-term, full-time contractor assignment lasting 5 weeks, requiring approximately 40 hours of work per week and at least 4 hours of daily overlap with PST.

Compensation follows a pay-per-task model, with $300 paid for each approved task that meets the specified quality and acceptance criteria. The average task is expected to require approximately 6–8 hours, although actual completion time may vary depending on complexity.

The contract does not include medical benefits or paid leave. Candidates should be available to begin immediately and commit to the required schedule throughout the engagement.

Selection Process

1. Interest Check Form: Complete the initial application and interest confirmation.
2. Technical Assessment: Complete a 120-minute assessment designed to evaluate relevant scientific and computational expertise.
3. Delivery Manager Review: Applications and assessment results will be reviewed before final selection.

Important application requirement: Candidates must use the same email address for the Interest Check Form and the agency portal submission.