Collins Odhiambo

Key Skills

Software

Appen

Humanatic

Mercor

Remotasks

Snorkel AI

Micro1

Data Annotation Tech

Top Subject Matter

Artificial Intelligence & Machine Learning

Cloud Computing / Internet Infrastructure

Enterprise Software / SaaS

Top Data Types

Video

Audio

Computer Code Programming

Top Label Types

Segmentation

Bounding Box

Classification

Object Detection

Text Generation

Transcription

Polygon

Freelancer Overview

With over 14 years of experience in AI, machine learning, and scalable software systems, I specialize in designing and deploying robust AI solutions that rely on high-quality data labeling and annotation. My expertise spans building multimodal data processing platforms that integrate text, image, and tabular data for predictive analytics, as well as leading teams in NLP, computer vision, and medical imaging projects. I have developed and optimized end-to-end ML pipelines using tools like TensorFlow, PyTorch, KubeFlow, and MLflow, ensuring data integrity and model accuracy at scale. My work includes training and fine-tuning models for legal document summarization, medical image diagnosis, and autonomous routing, all of which required meticulous management and curation of training datasets. I am passionate about building ethical, explainable AI systems and have a strong track record of collaborating with cross-functional teams to deliver impactful, data-driven solutions across healthcare, finance, and logistics domains.

Labeling Experience

ai annotation

AppenVideoSegmentation

ai annotate videos perfectly

ai annotate videos perfectly

2023 - 2024

AI/ML Model Training & Data Pipeline Engineer

Computer Code ProgrammingComputer Programming Coding

Led end-to-end AI model training and data labeling initiatives across healthcare, legal, and enterprise domains. At IBM Watson, directed NLP annotation pipelines for legal and medical document classification. At Quanata, built a multimodal data processing platform combining text, image, and tabular data for predictive analytics. Trained CNNs for medical imaging diagnosis achieving 92%+ precision, and developed BERT-based models fine-tuned on labeled legal datasets.

Led end-to-end AI model training and data labeling initiatives across healthcare, legal, and enterprise domains. At IBM Watson, directed NLP annotation pipelines for legal and medical document classification. At Quanata, built a multimodal data processing platform combining text, image, and tabular data for predictive analytics. Trained CNNs for medical imaging diagnosis achieving 92%+ precision, and developed BERT-based models fine-tuned on labeled legal datasets.

2016 - 2018

Education

Work History