For employers

Hire this AI Trainer

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

LoginCreate account
Invite to Job
L

Lovecase Singh

AI Training Expert | LLM Evaluation, Data Labeling & GenAI Systems

India flagBangalore, India
$50.00/hrIntermediateMercorScale AIRemotasks

Key Skills

Software

MercorMercor
Scale AIScale AI
RemotasksRemotasks
LabelboxLabelbox

Top Subject Matter

Generative AI
LLM QA evaluation
LLM fine-tuning

Top Data Types

TextText
ImageImage
DocumentDocument

Top Task Types

Fine-tuningFine-tuning
ClassificationClassification
Question AnsweringQuestion Answering
RLHFRLHF
Computer Programming/CodingComputer Programming/Coding

Freelancer Overview

I have hands-on experience in AI training data and evaluation through my work as a GenAI Intern and now Associate Software Engineer at o9 Solutions. During my internship, I built a Knowledge Bot Evaluation Framework using the RAGAs framework, where I designed structured evaluation pipelines to assess LLM outputs across dimensions like faithfulness, answer relevancy, context precision, and semantic similarity. This involved creating high-quality “golden datasets,” defining ground truths, and implementing systematic evaluation and rating workflows—directly aligning with AI training data processes such as labeling, scoring, and feedback generation. In addition, I have worked on fine-tuning LLMs using QLoRA with Hugging Face Transformers, gaining experience with training data preparation, prompt structuring, and model evaluation. My work also includes building RAG-based AI systems and agents, where I handle unstructured data, design retrieval pipelines, and improve response quality through iterative evaluation. I bring strong analytical skills, attention to detail, and experience working with structured data workflows, making me well-suited for AI training, evaluation, and data annotation tasks.

IntermediateEnglish

Labeling Experience

Gen-AI Intern, LLM Fine-tuning

TextFine Tuning
I fine-tuned a Large Language Model (LLM) using PEFT (QLoRA) on custom datasets targeted for internal QA. The process included data preparation, model adaptation, and performance benchmarking. This initiative advanced internal LLM capabilities for domain-specific question answering tasks.• Fine-tuned a language model using state-of-the-art parameter-efficient techniques. • Utilized Hugging Face Transformers, TRL (SFTTrainer), and BitsAndBytes for optimization. • Created and managed domain-appropriate training data for supervised model updating. • Benchmarked the adapted model on internal QA tasks for efficacy assessment.

I fine-tuned a Large Language Model (LLM) using PEFT (QLoRA) on custom datasets targeted for internal QA. The process included data preparation, model adaptation, and performance benchmarking. This initiative advanced internal LLM capabilities for domain-specific question answering tasks.• Fine-tuned a language model using state-of-the-art parameter-efficient techniques. • Utilized Hugging Face Transformers, TRL (SFTTrainer), and BitsAndBytes for optimization. • Created and managed domain-appropriate training data for supervised model updating. • Benchmarked the adapted model on internal QA tasks for efficacy assessment.

2024 - 2024

Gen-AI Intern, Knowledge Bot Evaluation

Text
As a Gen-AI Intern, I developed a Knowledge Bot Evaluation Framework to assess the response quality of AI-generated answers. I created a Golden Dataset with structured ground truths and established an evaluation pipeline for systematic quality checks. The framework was integrated with visualization tools and supported continuous feedback cycle improvements.• Evaluated knowledge bot outputs across metrics like Faithfulness, Answer Relevancy, and Semantic Similarity. • Constructed and validated a Golden Dataset to serve as ground truth for AI evaluation. • Employed tokenization and context retrieval techniques to support model analysis and compliance. • Deployed an interactive UI for metric visualization and feedback collection.

As a Gen-AI Intern, I developed a Knowledge Bot Evaluation Framework to assess the response quality of AI-generated answers. I created a Golden Dataset with structured ground truths and established an evaluation pipeline for systematic quality checks. The framework was integrated with visualization tools and supported continuous feedback cycle improvements.• Evaluated knowledge bot outputs across metrics like Faithfulness, Answer Relevancy, and Semantic Similarity. • Constructed and validated a Golden Dataset to serve as ground truth for AI evaluation. • Employed tokenization and context retrieval techniques to support model analysis and compliance. • Deployed an interactive UI for metric visualization and feedback collection.

2024 - 2024

Education

N

National Institute of Technology, Agartala

Bachelor of Technology, Computer Science and Engineering

Bachelor of Technology
2020 - 2024

Work History

O

o9 Solutions

Associate Software Engineer

Bangalore
2024 - Present
O

o9 Solutions

Gen-AI Intern

Bangalore
2024 - 2024