About the job
We are looking for a seasoned COBOL expert to join our team as a freelance consultant. In this role, you will collaborate with top AI model builders and enterprises to enhance the quality of AI training data projects. Your expertise will help establish success metrics, evaluate outputs, and provide strategic insights to accelerate project delivery. You will be integral in producing high-quality, domain-specific annotated datasets and code samples that adhere to the highest technical standards, with clearly defined deliverables and deadlines.
Key Responsibilities
Technical Standards, Quality Assurance, and Process Optimization
- Establish COBOL-specific quality metrics for dataset labeling and code annotation initiatives.
- Create project-specific standard operating procedures (SOPs), quality assurance rubrics, and reference materials to ensure compliance with client technical requirements.
- Analyze project outputs, including COBOL scripts and legacy modernization samples, verifying adherence to established standards and rectifying any defects prior to client submission.
- Conduct thorough quality assurance assessments on deliverables, tracking and resolving defects efficiently to meet project timelines.
- Provide detailed feedback to contractors, including remediation notes on COBOL syntax, logic, and legacy system patterns.
- Offer expert advice on tools, frameworks, emulators, and workflow enhancements to uphold quality benchmarks in mainframe and batch-processing environments.
- Manage specification changes and edge cases, drafting acceptance criteria or workarounds as necessary.
- Compile a library of exemplary COBOL code samples and dataset annotations to ensure consistency across projects.
Talent Evaluation and Output Enhancement
- Engage in technical assessments of contractor talent by reviewing COBOL code evaluations and task-based tests.
- Examine sample outputs from contractors and deliver actionable feedback to enhance code accuracy, clarity, and performance.
- Create tailored training resources, including:
- COBOL code quality standards (e.g., data division consistency, paragraph structuring)
- Best practices for writing clean, maintainable procedural code
- Reference materials for legacy system interaction
- Standards for COBOL dataset labeling

