Nupur

Key Skills

Software

Other

Top Subject Matter

LLM evaluation across healthcare

Travel Domain Expertise

Education Domain Expertise

Top Data Types

Text

Image

Top Task Types

Classification

Data Collection

Freelancer Overview

LLM Evaluation & Generation Specialist, Outlier. Brings 4+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Other, Telus AI, and Internal. Education includes Doctor of Pharmacy, Gujarat Technological University (2025). AI-training focus includes data types such as Text, Medical, and DICOM and labeling workflows including Evaluation, Rating, and Classification.

Labeling Experience

Executive – Pharmacovigilance & Clinical Research, Torrent Pharmaceuticals

Data Collection

Supported structured clinical safety data review processes requiring careful decision-making and documentation. Assisted with signal detection and case assessment workflows using established pharmacovigilance procedures. Ensured accuracy and organization of safety documentation to support downstream evaluation. • Reviewed and analyzed clinical safety data in a structured manner. • Supported documentation workflows for case assessment. • Contributed to signal detection and tracking activities. • Followed quality and decision-making processes for clinical data.

2025 - Present

AI Data Annotation & Response Evaluation, TELUS AI

TextClassification

Performed text, image, and video annotation using strict guidelines to support AI training and evaluation. Conducted output audits and flagged errors in model responses against defined expectations. Produced structured feedback notes to improve response quality and compliance with annotation standards. • Labeled multimodal data (text, image, video) per guideline instructions. • Verified outputs through audits and error flagging. • Documented evaluation notes and structured feedback for model iteration. • Ensured label consistency with gold/reference criteria.

2025 - 2025

LLM Evaluation & Generation Specialist, Outlier

OtherText

Designed multi-step, human-realistic scenarios and used them to evaluate LLM behavior. Applied scoring rubrics and produced gold-standard outputs to benchmark reasoning quality. Audited model performance and recorded reasoning failures across multiple domains including healthcare, travel, education, and fitness. • Authored and structured test cases/prompts for scenario-based evaluation. • Followed rubric-based scoring and comparison against gold standards. • Classified and documented error types and inconsistencies. • Generated evaluation notes on response quality and reasoning robustness.

2024 - 2025

Education

Work History