Debojyoti B.

Key Skills

Software

AWS SageMaker

Top Subject Matter

AI Tutoring Platform (Educational Content)

Educational Document Q&A

Top Data Types

Text

Document

Image

Top Task Types

Fine-tuning

Question Answering

Freelancer Overview

LLM Fine-tuning & Prompt Engineer (AI/ML Engineer at SlideCoach). Brings 4+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include AWS SageMaker, Internal, and Proprietary Tooling. Education includes Bachelor of Technology, Brainware University (2025). AI-training focus includes data types such as Text and Document and labeling workflows including Fine-tuning and Question Answering.

Labeling Experience

LLM Fine-tuning & Prompt Engineer (AI/ML Engineer at SlideCoach)

AWS SageMakerTextFine-tuning

Applied prompt engineering and performed LLM fine-tuning using domain-specific content to improve Q&A accuracy. Created and iteratively improved textual prompts and responses to enhance the performance of AI models on specific tutoring content. Worked on the end-to-end process of engineering datasets and providing supervised fine-tuning signals to large language models. • Used AWS Bedrock and SageMaker as the main platforms for LLM fine-tuning tasks. • Focused on prompt+response writing and question answering accuracy for educational slide-based material. • Automated aspects of data ingestion and model retraining for adaptive AI feedback loops. • Collaborated with tutoring subject matter experts to ensure data and labeling relevance.

2025 - Present

AI/ML Tech Intern - RAG-based Knowledge Labeling (Campus Ready)

DocumentQuestion Answering

Developed RAG-based systems that utilized labeled knowledge extracted from structured educational PDFs. Created and managed context windows and answer spans from annotated documents for question answering tasks. Validated and formatted labeled data to support retrieval-augmented generation for student-facing AI features. • Used ChromaDB and AWS Bedrock for data processing and RAG implementation. • Labeled answer keys, topics, and content sections from educational PDFs for automated Q&A. • Integrated labeled data into conversational AI workflows serving students. • Focused on structured document segmentation and context annotation for optimal knowledge retrieval.

2025 - 2025

Education

Work History