Ai Trainer
Current• Engineer high-quality prompts requesting insightful data analysis, manipulation, and visualization tasks reflective of real-world scenarios.• Evaluate prompts based on complexity, presence of code execution requirements, data accessibility, and expected output type.• Compose Python code solutions to prompts involving data importing, processing with libraries like Pandas/NumPy, and generating visualizations.• Rate code solutions on execution time, output accuracy, readability, inline documentation, performance optimization, safety, and design principles.• Assess responses of varying specificity degrees through multi-point rubrics covering clarity, accuracy, engagement, safety, formatting, and overall effectiveness.• Craft natural language responses based on code execution outputs in adherence to these rubrics.• Perform qualitative analysis to refine prompts, code solutions, and responses, iterating until achieving the highest scoring per rigorous criteria.• Identify and evaluate potential new data sources for relevance and quality to expand the dataset library.• Analyze the datasets for consistency, uniqueness, and adherence to patterns and rules.• Investigate and identify the underlying reasons for low-quality data and apply corrective actions.• Transform data from multiple sources across languages, formats, and structures to ensure compatibility with analysis tools.• Ensure the accuracy and integrity of data by applying checks and controls (deduplication, imputation, winsorizing, trimming, formatting, etc.).• Enhance quality during data collection by appending related information from external sources.• Ensure data handling complies with relevant laws, regulations, and ethical standards.• Write detailed error reports on inconsistencies and anomalies found during post-process audits.• Create comprehensive documentation for data sources, cleaning methodologies, and quality assurance processes to ensure data reliability and validity.