Aryan K.

Key Skills

Software

Appen

Telus

Top Subject Matter

No subject matter listed

Top Data Types

Document

Image

Text

Top Task Types

Data Collection

Evaluation/Rating

Prompt + Response Writing (SFT)

Question Answering

Text Generation

Freelancer Overview

I have extensive experience in AI training data preparation, labeling, and evaluation, working on multiple projects with TELUS Digital and CrowdGen AI (Appen ButlerHill). My work has included data structuring, collection, and quality validation for large language models (LLMs), as well as crafting and refining prompts to optimize model outputs. I have collaborated with cross-functional teams to ensure data accuracy, regulatory compliance, and consistency, contributing to high-performing AI-driven solutions.

Labeling Experience

Financial data processing for LLM

TelusTextText SummarizationData Collection

The project involved analyzing, structuring, and validating financial datasets to train and optimize LLMs for finance-specific applications. I worked on preparing large structured and unstructured financial data, including reports, risk assessment datasets, and compliance-related documents. A critical component was the integration of reasoning into dataset preparation and ensuring that the AI models could go beyond data retrieval to perform logical financial analysis, variance checks, and compliance reasoning. The datasets involved tens of thousands of financial records, ranging from numerical transactions to unstructured regulatory texts, which were transformed into training-ready formats.

2024 - 2025

LLM evlauation

AppenTextPrompt + Response Writing (SFT)

The project involved designing, labeling, and evaluating training datasets to enhance LLM performance. My responsibilities included crafting prompts, annotating responses, and conducting quality evaluations to ensure accuracy, contextual relevance, and compliance with project guidelines. I worked extensively with both structured and unstructured datasets, including financial documents, conversational text, and voice command data. The scope of the project was large-scale, encompassing tens of thousands of labeled and validated entries across multiple domains. A key part of my work was applying reasoning techniques during data labeling and prompt evaluation, ensuring the AI system could handle not only factual queries but also logical reasoning, financial problem-solving, and multi-step analysis.

2024 - 2025

Education

Work History