Avish K.

Key Skills

Software

CVAT

Data Annotation Tech

Top Subject Matter

General Machine Learning/Data Operations

Conversational AI/LLM Alignment

Legal Services & Contract Review

Top Data Types

Image

Text

Video

Document

Top Task Types

Audio Recording

Bounding Box

Prompt + Response Writing (SFT)

Question Answering

Text Summarization

Classification

RLHF

Freelancer Overview

ML Data Associate I - Data Annotation & Validation. Core strengths include Internal, Proprietary Tooling, and N. Education includes Bachelor of Technology, Godavari Institute of Engineering and Technology (2022). AI-training focus includes data types such as Text and labeling workflows including Classification and RLHF.

Labeling Experience

ML Data Associate I

TextClassification

As an ML Data Associate I at Amazon, I annotated and validated text, audio, image, and video data for machine learning model training. I collaborated closely with ML engineers to refine data quality, ensured adherence to project-specific guidelines, and consistently exceeded productivity and quality targets. I also maintained confidentiality and identified areas for process improvement to support high-quality AI development. • Annotated datasets across multiple data types using internal proprietary tools. • Reviewed, validated, and corrected labeled data for accuracy and consistency. • Provided constructive feedback for tool improvements and workflow optimizations. • Worked within the team to meet both quality assurance and daily throughput goals.

2024 - Present

Prompt Grader – AI Training Simulation

TextRLHF

During the MIL Prompt Grading Workflow – AI Training Simulation Project, I simulated Amazon's prompt grading pipeline to evaluate LLM-generated responses. I utilized MIL guidelines and detailed rubrics to annotate more than 500 prompt-response pairs for accuracy, fluency, completeness, and bias. These high-quality, QA-reviewed annotations contributed directly to RLHF and supervised LLM improvement. • Labeled hundreds of samples following structured MIL protocols for AI alignment. • Built and iterated datasets fit for Alexa, AWS Lex, and enterprise virtual assistants. • Used annotation tools and quality assurance workflows to uphold data standards. • Enhanced model safety and performance through human-in-the-loop feedback.

2023 - 2023

Education

Work History