Rutvik Waghmare

Key Skills

Software

Label Studio

CVAT

Data Annotation Tech

Mercor

OneForma

OpenCV AI Kit (OAK)

Scale AI

SuperAnnotate

Toloka

Top Subject Matter

Conversational AI

Speech/Conversational Data

Customer Conversation Text and Image Data

Top Data Types

Text

3D Sensor

Document

Top Task Types

Emotion Recognition

Bounding Box

Text Generation

RLHF

Prompt + Response Writing (SFT)

Text Summarization

Object Detection

Segmentation

Classification

Data Collection

Freelancer Overview

Expert AI Trainer / Generalist. Core strengths include Label Studio, Internal, and Proprietary Tooling. Education includes Bachelor of Technology, Netaji Subhas University of Technology (2023). AI-training focus includes data types such as Text and Audio and labeling workflows including Intent Tagging, Transcription, and Emotion Recognition.

Labeling Experience

QA Annotator & Prompt Writer

Label StudioDocumentRLHF

Personalization Search (p13n): Refining search model logic by evaluating the contextual relevance of personalized results for Large Language Models. Audio Agent Evals: Assessing the quality and performance of voice-based AI agents, focusing on intent recognition and natural language generation. RLHF Optimization: Continuously improving model reasoning through high-complexity Reinforcement Learning from Human Feedback.

Personalization Search (p13n): Refining search model logic by evaluating the contextual relevance of personalized results for Large Language Models. Audio Agent Evals: Assessing the quality and performance of voice-based AI agents, focusing on intent recognition and natural language generation. RLHF Optimization: Continuously improving model reasoning through high-complexity Reinforcement Learning from Human Feedback.

2023 - Present

Expert AI Trainer / Generalist

Label StudioText

In this role, I executed large language model (LLM) annotation workflows, placing a strong emphasis on intent tagging, summarization evaluation, and assessing response quality in conversational AI datasets. I conducted thorough data quality audits, rectifying missing and null values, and addressed edge cases to improve factual accuracy and safety. Working closely with machine learning teams, I ensured annotation standards were met, accelerating project onboarding while upholding strict confidentiality. • Implemented comprehensive LLM data annotation for conversational AI applications • Conducted 20% improvement in data quality and model safety by rigorously handling noisy, multi-speaker data • Focused on continual improvement of annotation guidelines via collaboration with ML teams • Maintained high standards of data integrity throughout the annotation process.

In this role, I executed large language model (LLM) annotation workflows, placing a strong emphasis on intent tagging, summarization evaluation, and assessing response quality in conversational AI datasets. I conducted thorough data quality audits, rectifying missing and null values, and addressed edge cases to improve factual accuracy and safety. Working closely with machine learning teams, I ensured annotation standards were met, accelerating project onboarding while upholding strict confidentiality. • Implemented comprehensive LLM data annotation for conversational AI applications • Conducted 20% improvement in data quality and model safety by rigorously handling noisy, multi-speaker data • Focused on continual improvement of annotation guidelines via collaboration with ML teams • Maintained high standards of data integrity throughout the annotation process.

2026 - Present

ML Data Operations Specialist

VideoComputer Programming Coding

AI/ML Data Associate | Amazon October 2024 – Present Ground Truth Data Generation: Execute high-precision annotation for various data modalities, including video, image, and text, to create the foundational "ground truth" used to train and fine-tune Amazon’s AI models. Complex Video Auditing: Perform frame-by-frame video analysis (e.g., stowing actions in fulfillment centers) using internal tools to identify specific activities, ensuring 99% accuracy in defect reduction and inventory tracking. LLM & Generative AI Training: Contribute to the responsible development of Large Language Models (LLMs) by generating factually accurate responses, ranking model outputs, and identifying logical inconsistencies or biases. Quality & Productivity Benchmarking: Maintain a consistent high-performance bar by meeting stringent daily targets for both quality (accuracy) and productivity (speed), often handling hundreds of data points per shift. Standard Operating Procedure (SOP) Optimization: Identify gaps or ambiguities in labeling guidelines and suggest improvements to SOPs and internal tools to unblock operations and enhance data fidelity. Root Cause Analysis: Analyze error patterns in datasets to identify the source of defects, proposing technical or process-oriented solutions to improve the overall machine learning lifecycle. Technical Scripting (Python/SQL): Utilize technical proficiency to automate repetitive validation tasks and manage data workflows within Linux/VM environments.

AI/ML Data Associate | Amazon October 2024 – Present Ground Truth Data Generation: Execute high-precision annotation for various data modalities, including video, image, and text, to create the foundational "ground truth" used to train and fine-tune Amazon’s AI models. Complex Video Auditing: Perform frame-by-frame video analysis (e.g., stowing actions in fulfillment centers) using internal tools to identify specific activities, ensuring 99% accuracy in defect reduction and inventory tracking. LLM & Generative AI Training: Contribute to the responsible development of Large Language Models (LLMs) by generating factually accurate responses, ranking model outputs, and identifying logical inconsistencies or biases. Quality & Productivity Benchmarking: Maintain a consistent high-performance bar by meeting stringent daily targets for both quality (accuracy) and productivity (speed), often handling hundreds of data points per shift. Standard Operating Procedure (SOP) Optimization: Identify gaps or ambiguities in labeling guidelines and suggest improvements to SOPs and internal tools to unblock operations and enhance data fidelity. Root Cause Analysis: Analyze error patterns in datasets to identify the source of defects, proposing technical or process-oriented solutions to improve the overall machine learning lifecycle. Technical Scripting (Python/SQL): Utilize technical proficiency to automate repetitive validation tasks and manage data workflows within Linux/VM environments.

2024 - Present

Education

Work History