For employers

Hire this AI Trainer

Sign in or create an account to invite AI Trainers to your job.

LoginCreate account
Invite to Job
Parth Darji

Parth Darji

Research Associate - Data Engineer (NLP Data Labeling)

CANADA flag
Windsor, Canada
$55.00/hrEntry LevelAws SagemakerAppenClickworker

Key Skills

Software

AWS SageMakerAWS SageMaker
AppenAppen
ClickworkerClickworker
LabelboxLabelbox
MercorMercor
MindriftMindrift
RemotasksRemotasks
SuperAnnotateSuperAnnotate
TolokaToloka
Scale AIScale AI
Other

Top Subject Matter

Legal Services & Contract Review
Regulatory Compliance & Risk Analysis
Legal Research & Document Analysis

Top Data Types

Computer Code ProgrammingComputer Code Programming
TextText
VideoVideo
DocumentDocument

Top Label Types

Classification
Computer Programming Coding
Entity Ner Classification
Prompt Response Writing SFT
Text Summarization

Freelancer Overview

Education includes Ontario College Graduate Certificate, Conestoga College (2020).

Entry LevelHindiEnglish

Labeling Experience

RWS - Data Annotation Contributor

VideoEvaluation Rating
Performed prompt/response evaluation and ranking tasks to improve large AI models. Assessed model outputs for accuracy, reasoning quality, and safety compliance. Flagged hallucinations and incorrect responses using structured evaluation rubrics.

Performed prompt/response evaluation and ranking tasks to improve large AI models. Assessed model outputs for accuracy, reasoning quality, and safety compliance. Flagged hallucinations and incorrect responses using structured evaluation rubrics.

2025 - Present

DevDocBuddy Project – AI Documentation Assistant (Text Summarization Labeling)

OtherTextText Summarization
I developed an AI-driven documentation assistant capable of converting complex technical documents into plain-English summaries. My responsibilities included integrating large language models for summarization and question-answering tasks, as well as chunk-level dataset curation and annotation. This project required producing reliable source-cited answers for end users by curating, summarizing, and labeling document text data. • Used LangChain and Hugging Face models for text summarization and Q&A • Curated and processed PDFs, Markdown files, and READMEs for AI training • Implemented and evaluated plain-English summaries with cited sources • Annotated and validated output for dataset reliability and debuggability.

I developed an AI-driven documentation assistant capable of converting complex technical documents into plain-English summaries. My responsibilities included integrating large language models for summarization and question-answering tasks, as well as chunk-level dataset curation and annotation. This project required producing reliable source-cited answers for end users by curating, summarizing, and labeling document text data. • Used LangChain and Hugging Face models for text summarization and Q&A • Curated and processed PDFs, Markdown files, and READMEs for AI training • Implemented and evaluated plain-English summaries with cited sources • Annotated and validated output for dataset reliability and debuggability.

2023 - Present

Alignerr — AI Training & Evaluation Tasks

TextRLHF
Completed AI training tasks focused on prompt evaluation and response quality assessment for large language models. Reviewed generated answers to ensure compliance with safety, factual accuracy, and instruction adherence guidelines. Compared multiple AI responses and ranked them based on logical reasoning, completeness, and relevance.

Completed AI training tasks focused on prompt evaluation and response quality assessment for large language models. Reviewed generated answers to ensure compliance with safety, factual accuracy, and instruction adherence guidelines. Compared multiple AI responses and ranked them based on logical reasoning, completeness, and relevance.

2025 - 2025

Instagram video product tagging (commerce metadata)

OtherVideoEntity Ner Classification
Tagged products shown in short-form Instagram videos by identifying the correct item(s), brand/category, and key attributes (e.g., color, style, material, pattern, size) and linking them to the best-matching catalog entry.

Tagged products shown in short-form Instagram videos by identifying the correct item(s), brand/category, and key attributes (e.g., color, style, material, pattern, size) and linking them to the best-matching catalog entry.

2025 - 2025

Outlier - LLM Prompt/Response Evaluation & Improvement

Computer Code ProgrammingComputer Programming Coding
Identified common failure patterns such as hallucinations, incomplete reasoning, or poor formatting. Provided improved responses or rewritten outputs that better followed prompt instructions and improved answer quality. Followed project rubrics and evaluation frameworks to score outputs across categories such as helpfulness, correctness, clarity, and safety.

Identified common failure patterns such as hallucinations, incomplete reasoning, or poor formatting. Provided improved responses or rewritten outputs that better followed prompt instructions and improved answer quality. Followed project rubrics and evaluation frameworks to score outputs across categories such as helpfulness, correctness, clarity, and safety.

2024 - 2025

Education

C

Conestoga College

Ontario College Graduate Certificate, Computer Science

Ontario College Graduate Certificate
2019 - 2020
C

Conestoga College

Ontario College Graduate Certificate, Big Data Solutions Architecture

Ontario College Graduate Certificate
2019 - 2020

Work History

I

IBM

Full Stack Cloud Software Developer

Toronto
2022 - Present
I

IBM Canada

Full Stack Cloud Software Developer

Toronto
2022 - Present