Talent Dev

Key Skills

Software

CVAT

Top Subject Matter

No subject matter listed

Top Data Types

Image

Top Label Types

Bounding Box

Segmentation

Freelancer Overview

I am a Senior AI Engineer with extensive experience in building and deploying advanced AI systems, specializing in NLP, Conversational AI, and large language models. My background includes designing and fine-tuning LLMs using domain-specific datasets, implementing Retrieval-Augmented Generation (RAG) pipelines, and developing data pipelines for entity extraction, intent recognition, and sentiment analysis. I have hands-on expertise with data annotation, prompt engineering, and automated data enrichment workflows, leveraging tools like LangChain, LlamaIndex, Hugging Face Transformers, and various vector databases. My work spans the full AI lifecycle, from data labeling and quality scoring to model training and evaluation, ensuring high-quality, context-aware responses in enterprise environments. I am adept at bridging the gap between raw data and actionable AI insights, always prioritizing data quality and operational efficiency.

Labeling Experience

Large-Scale Multimodal Data Annotation for Autonomous Driving Pipeline

CVATImageBounding BoxSegmentation

Performed high-volume data annotation for an autonomous vehicle perception pipeline, completing 20,000+ image annotations across urban, suburban, and highway driving scenarios. Tasks included 2D/3D bounding box annotation, semantic segmentation, lane marking, and polygon annotation for 40+ object classes including vehicles, pedestrians, cyclists, traffic signs, and road markings. Consistently maintained an annotation accuracy rate above 97% through adherence to strict labeling guidelines and multi-stage QA review processes. Skilled in handling complex edge cases such as heavy occlusion, low-light environments, and adverse weather conditions.

Performed high-volume data annotation for an autonomous vehicle perception pipeline, completing 20,000+ image annotations across urban, suburban, and highway driving scenarios. Tasks included 2D/3D bounding box annotation, semantic segmentation, lane marking, and polygon annotation for 40+ object classes including vehicles, pedestrians, cyclists, traffic signs, and road markings. Consistently maintained an annotation accuracy rate above 97% through adherence to strict labeling guidelines and multi-stage QA review processes. Skilled in handling complex edge cases such as heavy occlusion, low-light environments, and adverse weather conditions.

2022 - 2025

Education

Work History