For employers

Hire this AI Trainer

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

Invite to Job
Sukant Kumar

Sukant Kumar

AI Data Trainer (1.5 Year) | Web Development | Coder

INDIA flag
Chhapra, India
$30.00/hrIntermediateOther

Key Skills

Software

Other

Top Subject Matter

No subject matter listed

Top Data Types

Computer Code ProgrammingComputer Code Programming
DocumentDocument
TextText

Top Label Types

Action Recognition
Computer Programming Coding
Evaluation Rating
Text Generation
Text Summarization

Freelancer Overview

I have extensive experience in AI training and data labeling, particularly in the fields of emotion recognition, NLP, and response evaluation. I have worked as an AI Data Trainer at Soul AI, where I was responsible for evaluating AI-generated responses, comparing multiple options, and providing detailed feedback on which responses were better, including assessments of hedging and accuracy. I also played a key role in fine-tuning AI models, ensuring their outputs were optimized for clarity and relevance. In addition, I am currently working on projects involving OpenAI through Outlier, where I apply my expertise in AI feedback, model evaluation, and coding. These experiences have allowed me to work on cutting-edge projects, such as NLP model evaluation, text generation, and summarization, further sharpening my skills in data labeling, model optimization, and AI system evaluation.

IntermediateHindiEnglish

Labeling Experience

AI Data Trainer | Fact-Checking & Response Evaluation

OtherTextQuestion AnsweringText Generation
In this role at Soul AI, I focus on fact-checking AI responses to ensure their accuracy and reliability. The project involves reviewing a prompt and determining whether the AI-generated response is understandable. Once I confirm that, I then assess two potential responses from the AI. I analyze them for factual accuracy, checking if the claims made in the responses are supported by credible, verifiable sources. The evaluation includes checking for hedging (where the AI avoids committing to a definitive answer), canned responses (where the model gives vague, unhelpful answers), and the overall clarity and completeness of the responses. My task also involves ranking the responses based on their factual accuracy and explaining the rationale behind my choices, using reliable online sources to back up my findings. Additionally, I rate the accuracy of each response on a scale from “Accurate” to “Inaccurate,” providing detailed feedback and sources used for verification.

In this role at Soul AI, I focus on fact-checking AI responses to ensure their accuracy and reliability. The project involves reviewing a prompt and determining whether the AI-generated response is understandable. Once I confirm that, I then assess two potential responses from the AI. I analyze them for factual accuracy, checking if the claims made in the responses are supported by credible, verifiable sources. The evaluation includes checking for hedging (where the AI avoids committing to a definitive answer), canned responses (where the model gives vague, unhelpful answers), and the overall clarity and completeness of the responses. My task also involves ranking the responses based on their factual accuracy and explaining the rationale behind my choices, using reliable online sources to back up my findings. Additionally, I rate the accuracy of each response on a scale from “Accurate” to “Inaccurate,” providing detailed feedback and sources used for verification.

2024

AI Data Trainer | Response Evaluation & NLP Optimization & coding tasks

OtherTextQuestion AnsweringText Generation
I am currently working as an AI Data Trainer with OpenAI through their platform, having been selected through Outlier. In this role, I evaluate user-assistant responses generated by OpenAI models. My tasks include reviewing two different responses to a given prompt and selecting the better response based on quality, clarity, correctness, and relevance. This evaluation process involves tasks like coding-related questions (JavaScript, SQL, web development), as well as more general queries (math, logic, etc.). I assess the quality of the responses, identifying issues like hedging and providing reasoning for why one response is better than another. Additionally, I contribute to fine-tuning the AI models by providing feedback and ratings, which helps improve the overall accuracy and performance of the AI systems.

I am currently working as an AI Data Trainer with OpenAI through their platform, having been selected through Outlier. In this role, I evaluate user-assistant responses generated by OpenAI models. My tasks include reviewing two different responses to a given prompt and selecting the better response based on quality, clarity, correctness, and relevance. This evaluation process involves tasks like coding-related questions (JavaScript, SQL, web development), as well as more general queries (math, logic, etc.). I assess the quality of the responses, identifying issues like hedging and providing reasoning for why one response is better than another. Additionally, I contribute to fine-tuning the AI models by providing feedback and ratings, which helps improve the overall accuracy and performance of the AI systems.

2024

Education

C

Chandigarh university

Bachelor in Computer Science, Computer Science

Bachelor in Computer Science
2020 - 2024

Work History

S

Soul AI

AI Data Trainer

Remote
2024 - Present
O

OpenAI (via Outlier)

AI Data Trainer

Remote
2024 - Present