Ken Langat

Key Skills

Software

CVAT

Top Subject Matter

No subject matter listed

Top Data Types

3D Sensor

Top Label Types

Bounding Box

Polygon

Segmentation

Cuboid

Text Generation

Freelancer Overview

I am an experienced AI Training Specialist with over 5 years of hands-on expertise in data annotation, model evaluation, and quality assurance for large-scale machine learning and generative AI systems. My work spans NLP, computer vision, and multimodal domains, where I have contributed to the development and refinement of production-grade LLMs through advanced annotation, prompt engineering, RLHF workflows, and preference ranking. I excel at designing labeling guidelines, conducting rigorous dataset reviews, and collaborating with cross-functional teams to ensure high-quality, reliable training data. Proficient with tools like Scale, Labelbox, Appen, and Python, I am committed to improving model accuracy, safety, and alignment through meticulous human-in-the-loop processes and structured feedback.

Labeling Experience

Senior AI Training Specialist

CVAT3D SensorBounding BoxPolygon

I’ve been working as a Senior AI Training Specialist on large-scale data annotation projects that support both computer vision and NLP model development. My day-to-day work includes labeling and reviewing multimodal datasets such as 3D sensor data, images, and text using CVAT, where I handle tasks like bounding boxes, polygons, key points, segmentation, object detection, classification, tracking, and RLHF evaluations. I regularly manage high-volume workloads while following detailed guidelines and quality checks to make sure the data is accurate and model-ready. I also collaborate closely with QA teams to resolve edge cases, refine labeling standards, and improve workflows, helping ensure the datasets are reliable for training and fine-tuning production-level AI systems.

I’ve been working as a Senior AI Training Specialist on large-scale data annotation projects that support both computer vision and NLP model development. My day-to-day work includes labeling and reviewing multimodal datasets such as 3D sensor data, images, and text using CVAT, where I handle tasks like bounding boxes, polygons, key points, segmentation, object detection, classification, tracking, and RLHF evaluations. I regularly manage high-volume workloads while following detailed guidelines and quality checks to make sure the data is accurate and model-ready. I also collaborate closely with QA teams to resolve edge cases, refine labeling standards, and improve workflows, helping ensure the datasets are reliable for training and fine-tuning production-level AI systems.

2021

Education

Work History