Pranay Jain

Key Skills

Software

Snorkel AI

Top Subject Matter

AI Benchmarking

Large Language Models

Code Generation

Top Data Types

Text

Document

Top Task Types

Evaluation Rating

Prompt Response Writing SFT

Computer Programming Coding

Red Teaming

Text Generation

RLHF

Question Answering

Fine Tuning

Freelancer Overview

Final-year B.Tech student who is analytical and pays attention to detail. They have worked with data processing, AI model evaluation, and structured problem-solving. Worked with real-world datasets in Python and SQL before, with a strong focus on data quality, finding patterns, and logical reasoning. Good at finding mistakes, dealing with edge cases, and making sure that structured data workflows are correct. I'm currently an AI Training and Evaluation Contributor, where I make high-quality reasoning tasks, look at LLM outputs, and give detailed feedback to help the model work better. Want to learn more about data annotation, AI training, and how to make models more accurate by giving them precise and well-organized data.

Labeling Experience

AI Training & Evaluation Contributor (Contract)

Snorkel AIText

As an AI Training & Evaluation Contributor for Xelron AI, I designed and assessed challenging tasks for large language models as part of benchmarking and evaluation projects. My work involved creating adversarial prompts, conducting detailed response evaluations, and providing structured feedback on advanced model reasoning. I contributed to both natural language and code-based tasks, supporting improvements in both general and technical AI accuracy. • Developed high-difficulty reasoning benchmarks to test LLM capabilities • Evaluated model outputs for correctness, depth, and consistency • Provided structured human feedback to identify strengths and errors in reasoning • Compared AI models on performance metrics such as code logic and documentation quality.

As an AI Training & Evaluation Contributor for Xelron AI, I designed and assessed challenging tasks for large language models as part of benchmarking and evaluation projects. My work involved creating adversarial prompts, conducting detailed response evaluations, and providing structured feedback on advanced model reasoning. I contributed to both natural language and code-based tasks, supporting improvements in both general and technical AI accuracy. • Developed high-difficulty reasoning benchmarks to test LLM capabilities • Evaluated model outputs for correctness, depth, and consistency • Provided structured human feedback to identify strengths and errors in reasoning • Compared AI models on performance metrics such as code logic and documentation quality.

2026 - Present

Education

Work History