Desmond C.

Key Skills

Software

Other

Mindrift

Toloka

Top Subject Matter

AI-generated video and motion graphics for marketing/education content

Educational video production

Top Data Types

Video

Image

Audio

Top Task Types

Classification

Object Detection

Fine-tuning

Data Collection

Transcription

Question Answering

Text Generation

Freelancer Overview

AI image and video creator — Cosmic Data Studios. Brings 6+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Other. Education includes Certified Video Creator and Sound Engineer, Cyber World College (2020). AI-training focus includes data types such as Video and labeling workflows including Classification.

Labeling Experience

Large Language Model Instruction Fine-Tuning Dataset

TextFine-tuning

Curate and annotate a high-quality dataset of instruction-response pairs for fine-tuning a conversational Ai assistant across domains including coding, creative writing, summarization, and Q&A TASKS PERFORMED 1, Wrote and revised 2,500 original instruction prompts with gold- standard responses 2, Ranked multiple candidate responses using pairwise preference scoring 1-7 scale 3, Flagged harmful, biased content for removal PROJECT SIZE; 12,000 instruction-response pairs; team of 5 annotators

2025 - 2025

AI image and video creator — Cosmic Data Studios

OtherVideoClassification

Created and produced AI image and video content for online platforms, applying creative transformations and ensuring outputs met studio quality standards. Worked from concept through final production, coordinating with a creative team to improve turnaround times and client satisfaction. Trained a team of junior editors to maintain consistent quality while enhancing creativity across video outputs. • Produced and edited 40 promotional videos for online distribution. • Improved post-production workflow to reduce editing times by 30%. • Introduced motion-graphics techniques that increased engagement by 40%. • Collaborated with marketing to enhance video narratives, increasing viewership by 15%.

2023 - 2025

Computer Vision Annotation For Autonomous Vehicle Object detection

ImageObject Detection

Annotate street level images to train an object detection model for identifying vehicles, pedestrians, cyclists and traffic signs under varied lighting and weather conditions project size; 45,000 images ; team of 8 annotators 1,Drew bounding around 15 objects classes using polygon and cuboid annotation techniques 2, Applied attribute tagging e.g occlusion level, truncation and weather condition 3, Reviewed edge cases such as partially visible objects and low contrast night images

2023 - 2023

Education

Work History