Divyansh Pandey

Key Skills

Software

Label Studio

Top Subject Matter

No subject matter listed

Top Data Types

Audio

Image

Medical Dicom

Text

Video

Top Label Types

Action Recognition

Audio Recording

Bounding Box

Classification

Data Collection

Emotion Recognition

Entity Ner Classification

Evaluation Rating

Fine Tuning

Mapping

Object Detection

Polygon

Question Answering

Transcription

Freelancer Overview

I am a Computer Science Engineering student specializing in AI and Machine Learning, with hands-on experience in building and optimizing data-driven AI systems. My background includes developing decentralized federated learning models for medical image classification, where I addressed data heterogeneity and ensured strict data privacy, as well as engineering robust ML pipelines for real-time fraud detection and semantic search applications. I am skilled in data annotation, feature engineering, and managing complex multimodal datasets, leveraging tools such as Python, PyTorch, TensorFlow, Hugging Face, and vector databases like FAISS and Pinecone. My work spans computer vision, NLP, and time series domains, and I have a strong foundation in deploying scalable, production-grade AI solutions with a focus on data quality, observability, and system reliability.

Labeling Experience

Social Winter of code problem: Animal-info-provider-chatbot

Label StudioImageBounding BoxPolygon

This project is a Flask-based application designed to integrate multiple components for handling imagery data, classifying animals, and providing detailed information about them through a custom QnA chatbot. The system uses MongoDB for storage, OpenCV for image classification, and a custom RAG (Retrieval-Augmented Generation) model for answering questions about animals based on a custom dataset.

This project is a Flask-based application designed to integrate multiple components for handling imagery data, classifying animals, and providing detailed information about them through a custom QnA chatbot. The system uses MongoDB for storage, OpenCV for image classification, and a custom RAG (Retrieval-Augmented Generation) model for answering questions about animals based on a custom dataset.

2024 - 2025

Education

Work History