Mansi K.

Key Skills

Software

SuperAnnotate

Mercor

Micro1

Labelbox

Internal/Proprietary Tooling

Top Subject Matter

Large Language Model (LLM) evaluation

Responsible AI

Multimodal AI systems

Top Data Types

Text

Image

Video

Top Task Types

Classification

Bounding Box

Point/Key Point

Object Detection

Text Generation

RLHF

Red Teaming

Evaluation/Rating

Segmentation

Freelancer Overview

Machine Learning / Responsible AI Analyst. Brings 2+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Internal and Proprietary Tooling. Education includes Bachelor of Technology, Sai University (2022). AI-training focus includes data types such as Text and labeling workflows including Evaluation, Rating, and Classification.

Labeling Experience

Data Annotator — Multimodal AI Systems

TextClassification

This role comprised annotation and validation of multimodal datasets for training and evaluation of enterprise-scale AI models. Emphasis was placed on maintaining high annotation quality standards to support model alignment and dataset validation. The position supported workflows for AI robustness across several concurrent projects. • Annotated text, image, and structured data for AI model training. • Validated data to ensure alignment and robustness in outputs. • Supported multiple annotation projects with strict quality measures. • Facilitated collaboration between annotation and engineering teams.

2025 - Present

Machine Learning / Responsible AI Analyst

Text

This experience involved conducting structured evaluations of large language models using red-team and blue-team methodologies. The focus was on assessing model safety, hallucination, alignment, and response quality to improve production AI systems. Policy-aligned prompts were engineered for benchmarking, and findings informed model fine-tuning and risk mitigation strategies. • Conducted adversarial testing and model evaluation for LLM safety and alignment. • Designed prompts for bias, misinformation, and robustness assessment. • Maintained scalable AI quality pipelines to improve output consistency. • Documented failure cases and analysis for ML team collaboration.

2025 - Present

Education

Work History