Ange Melissa Ishimwe

Key Skills

Software

Appen

Clickworker

Toloka

Other

Top Subject Matter

No subject matter listed

Top Data Types

Audio

Document

Image

Top Task Types

Action Recognition

Classification

Data Collection

Question Answering

Text Generation

Freelancer Overview

I have hands-on experience in data analysis and AI training through my work at MeshPower, where I manage three major software initiatives. I analyze raw electrical data, including voltage levels, BMS readings, and system health, to generate actionable insights for dashboards and client presentations. I also contribute to piloting a data collection software for solar systems, which forms the foundation for an interactive AI aimed at enabling easy diagnostics and system understanding for end users. This work requires precise technical knowledge, attention to detail, and the ability to translate complex engineering data into clear, actionable information. Beyond data handling, I am responsible for pitching software ideas and solutions, bridging technical insights with business and social impact goals. My background in Energy Engineering, combined with deep familiarity with electrical terminology, AI concepts, and software systems, equips me to create high-quality prompts, solutions, and evaluation rubrics. I am comfortable ensuring technical accuracy, calibrating difficulty, and aligning AI outputs with real-world engineering contexts, making me highly suited for AI training data roles.

Labeling Experience

Project Whatcom

AppenImagePolygonPoint Key Point

As part of the WHATCOM project, I uded methods such as polygon, bounding box, and semantic segmentation to label images and video frames accurately. I annotated objects including vehicles, pedestrians, traffic signs, and road markings to train and improve computer vision models for navigation and safety systems. This work required strict attention to detail, consistency across frames, and adherence to labeling guidelines to ensure high-quality datasets for AI training and model validation.

As part of the WHATCOM project, I uded methods such as polygon, bounding box, and semantic segmentation to label images and video frames accurately. I annotated objects including vehicles, pedestrians, traffic signs, and road markings to train and improve computer vision models for navigation and safety systems. This work required strict attention to detail, consistency across frames, and adherence to labeling guidelines to ensure high-quality datasets for AI training and model validation.

2023 - 2024

Education

Work History