Drashti Chovatiya

Evaluated 500+ AI-generated responses for accuracy, relevance, bias, and quality, contributing to continuous LLM performance improvement. Designed a multi-dimensional evaluation framework covering accuracy, relevance, bias, tone, and completeness to standardize AI output assessment.

Designed and implemented AI-powered ERP automation by integrating Claude Code with BuildSmart, reducing a 2-hour daily manual process to a 15-minute workflow and saving 16+ hours/week per coordinator. Architected standardized ERP workflows across departments, cutting manual data entry ~40% and compressing month-end close from 5 days to 2.5.

Executed AI evaluation workflows including prompt design, response validation, quality assurance, and multilingual output review to improve model performance. Managed end-to-end AI project operations using Airtable, Slack, and platform dashboards for project tracking, workflow coordination, and issue resolution.

Delivered market research and competitor analysis across FMCG and consumer sectors, producing actionable insights through quantitative and qualitative research. Coordinated global consulting projects by managing research deliverables, client documentation, and cross-functional communication while ensuring timely project execution.

Pursued a Bachelor of Business Administration in Finance, gaining a strong foundation in financial principles and business management.