Kuma T.

Key Skills

Software

Other

Data Annotation Tech

Top Subject Matter

Healthcare AI / disease outbreak prediction

Data structuring for event management analytics

Top Data Types

Computer Code Programming

Top Task Types

Fine-tuning

Data Collection

Freelancer Overview

HealthySync — Healthcare AI system (model training and fine-tuning). Brings 4+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Other. Education includes Bachelor's Degree, Addis Ababa Science and Technology University (2024) and Software Development and Problem Solving, African to Silicon Valley (A2SV) (2024). AI-training focus includes data types such as Medical, DICOM, and Computer Code and labeling workflows including Fine-tuning and Data Collection.

Labeling Experience

HealthySync — Healthcare AI system (model training and fine-tuning)

OtherFine-tuning

Created and led end-to-end development of a healthcare AI system for disease outbreak prediction and accuracy improvement. Trained and fine-tuned a Facebook Prophet model using 45,000+ cleaned data points derived from the project’s dataset. Built supporting pipelines to improve model robustness for real-world variability in the training data. • Focused on disease-outbreak prediction as the primary AI training task • Performed model training and fine-tuning with cleaned historical datapoints • Implemented feature engineering to handle missing data and seasonal trends • Validated impact with reported prediction metrics (e.g., MAPE)

2024 - 2025

Back-End Developer — Digital Event Management System (data organization to support analytics)

OtherData Collection

Developed a Digital Event Management System to improve event management efficiency for university events. While the role was primarily software engineering, it supported structured event data handling across 100+ events and replaced manual processes with a digital workflow. The work enabled consistent capture and organization of event-related records that can be used for downstream analytics. • Built systems to streamline and standardize event data flows • Replaced manual event tracking with a digitized process • Improved processing efficiency by 90% for recurring university events • Supported scalable usage across 100+ university events

2023 - 2023

Education

Work History