Himanshu S.

Key Skills

Software

Other

Top Subject Matter

LLM response evaluation

data annotation

dataset validation

Top Data Types

Text

Computer Code Programming

Top Task Types

Prompt + Response Writing (SFT)

Entity (NER) Classification

Freelancer Overview

AI Model Evaluation & Data Annotation Projects (Freelance / Self Learning). Core strengths include Other. Education includes Bachelor of Engineering, UIET, Panjab University, Chandigarh (2026). AI-training focus includes data types such as Text, Computer Code, and Programming and labeling workflows including Evaluation, Rating, and Prompt + Response Writing (SFT).

Labeling Experience

AI Model Evaluation & Data Annotation Projects

Entity (NER) Classification

No description provided.

2024 - Present

Educational Content Dataset Creation & AI Response Evaluation (Projects)

OtherPrompt + Response Writing (SFT)

Work included creating and preparing educational data for AI-assisted learning systems through structured labeling and validation. I built educational datasets intended for downstream model training by ensuring consistency and accuracy. I also supported LLM prompt testing by ranking and improving generated outputs based on evaluation criteria. • Created labeled educational datasets for AI training • Ensured dataset quality, consistency, and accuracy • Evaluated and ranked LLM outputs by correctness and adherence • Identified hallucinations and suggested improvements for model performance

2024 - Present

AI Model Evaluation & Data Annotation Projects (Freelance / Self Learning)

OtherText

Freelance work focused on evaluating AI-generated responses and performing dataset labeling and validation for AI training workflows. I assessed outputs for factual accuracy, instruction following, and overall quality while identifying hallucinations and inconsistencies. I also applied prompt engineering to improve response quality and reliability. • Evaluated AI responses across multiple quality dimensions • Labeled and validated dataset content for training readiness • Compared model outputs to detect contradictions and errors • Used prompt engineering to refine generation performance

2024 - Present

Education

Work History