Arpan Das

Key Skills

Software

Scale AI

Appen

Google Cloud Vertex AI

Labelbox

Top Subject Matter

No subject matter listed

Top Data Types

Computer Code Programming

Text

Top Task Types

Question Answering

RLHF

Evaluation Rating

Computer Programming Coding

Prompt Response Writing SFT

Classification

Translation Localization

Freelancer Overview

Student Researcher. Brings 4+ years of professional experience across complex professional workflows, research, and quality-focused execution.

Labeling Experience

Quality Control (QC) Expert – Biology Domain

Google Cloud Vertex AITextEvaluation Rating

At Deccan AI as a Quality Control (QC) Expert in the Biology Domain, I reviewed and corrected biology Q&A pairs for benchmarking LLMs. I ensured technical and conceptual accuracy and evaluated SME submissions using defined QC rubrics. My work contributed to refining QC and dataset annotation workflows for generative AI model training. • Checked step-by-step solutions for correctness and clarity. • Ensured rigorous academic standards and formatting using LaTeX. • Provided feedback and justifications for corrections. • Helped enhance annotation guidelines and dataset quality.

At Deccan AI as a Quality Control (QC) Expert in the Biology Domain, I reviewed and corrected biology Q&A pairs for benchmarking LLMs. I ensured technical and conceptual accuracy and evaluated SME submissions using defined QC rubrics. My work contributed to refining QC and dataset annotation workflows for generative AI model training. • Checked step-by-step solutions for correctness and clarity. • Ensured rigorous academic standards and formatting using LaTeX. • Provided feedback and justifications for corrections. • Helped enhance annotation guidelines and dataset quality.

2025

LLM Evaluation & Prompt Development Specialist

AppenTextEvaluation Rating

As an LLM Evaluation & Prompt Development Specialist at CrowdGen (Appen AI), I created complex questions to evaluate large language models' comprehension and capabilities. My work involved ranking AI-generated responses and designing nuanced prompts to test LLM handling of culturally specific, context-heavy content. I maintained high accuracy standards and performed in-depth response analysis to detect hallucinations and logical gaps. • Developed questions and prompts targeting reasoning and localization ability. • Applied strict QA and context-based evaluation benchmarks. • Identified weaknesses and improved LLM reasoning through detailed analysis. • Ensured evaluation speed and quality within structured benchmarks.

As an LLM Evaluation & Prompt Development Specialist at CrowdGen (Appen AI), I created complex questions to evaluate large language models' comprehension and capabilities. My work involved ranking AI-generated responses and designing nuanced prompts to test LLM handling of culturally specific, context-heavy content. I maintained high accuracy standards and performed in-depth response analysis to detect hallucinations and logical gaps. • Developed questions and prompts targeting reasoning and localization ability. • Applied strict QA and context-based evaluation benchmarks. • Identified weaknesses and improved LLM reasoning through detailed analysis. • Ensured evaluation speed and quality within structured benchmarks.

2025

AI Trainer

LabelboxTextClassification

As an AI Trainer at Aligneer, I focused on training and refining AI language models with English language data. I labeled, curated, and developed high-quality datasets to improve model performance, automating annotation and quality workflows. Through collaboration with technical teams, I improved annotation tool usability and helped fine-tune generative AI models via data-driven strategies. • Handled contextual annotation and language data processing. • Automated and streamlined data labeling and checks. • Enhanced model performance through curated training data. • Supported tool improvement and workflow optimization.

As an AI Trainer at Aligneer, I focused on training and refining AI language models with English language data. I labeled, curated, and developed high-quality datasets to improve model performance, automating annotation and quality workflows. Through collaboration with technical teams, I improved annotation tool usability and helped fine-tune generative AI models via data-driven strategies. • Handled contextual annotation and language data processing. • Automated and streamlined data labeling and checks. • Enhanced model performance through curated training data. • Supported tool improvement and workflow optimization.

2024

Social Media Ad Rater

AppenTextEvaluation Rating

As a Social Media Ad Rater for CrowdGen (Appen AI), I evaluated and rated the relevance, accuracy, and appropriateness of social media ads across various categories. My tasks included daily reviews, identifying misleading content, and providing high-quality feedback to improve ad targeting algorithms. I maintained flexibility and professionalism while handling sensitive material remotely. • Rated ad sets based on platform guidelines and quality measures. • Flagged content for misleading or inappropriate material. • Provided insights into public engagement and sentiment. • Ensured consistent quality and adherence to time constraints.

As a Social Media Ad Rater for CrowdGen (Appen AI), I evaluated and rated the relevance, accuracy, and appropriateness of social media ads across various categories. My tasks included daily reviews, identifying misleading content, and providing high-quality feedback to improve ad targeting algorithms. I maintained flexibility and professionalism while handling sensitive material remotely. • Rated ad sets based on platform guidelines and quality measures. • Flagged content for misleading or inappropriate material. • Provided insights into public engagement and sentiment. • Ensured consistent quality and adherence to time constraints.

2024

Multilingual LLM Evaluation and Biology Prompt Annotation

Scale AIComputer Code ProgrammingQuestion AnsweringRLHF

Worked on multiple high-impact AI training projects involving multilingual prompt writing, question answering, and LLM evaluation with a domain focus on biology. Tasks included generating and evaluating AI model outputs, correcting scientific inaccuracies, labeling biological entities, and assessing prompt-response quality using detailed rubrics. Ensured LaTeX formatting for complex academic content. Contributed to large-scale dataset creation across English, Hindi, and Bengali for training and fine-tuning state-of-the-art generative language models. Projects emphasized quality benchmarks, rubric-based scoring, and iterative feedback integration.

Worked on multiple high-impact AI training projects involving multilingual prompt writing, question answering, and LLM evaluation with a domain focus on biology. Tasks included generating and evaluating AI model outputs, correcting scientific inaccuracies, labeling biological entities, and assessing prompt-response quality using detailed rubrics. Ensured LaTeX formatting for complex academic content. Contributed to large-scale dataset creation across English, Hindi, and Bengali for training and fine-tuning state-of-the-art generative language models. Projects emphasized quality benchmarks, rubric-based scoring, and iterative feedback integration.

2023

Show more

Education

Work History