Open to opportunities

Luis Okech

@luisokech

Message

Machine Learning Engineer and AI Trainer focused on model optimization, fairness, and NLP. Data Annotator, video

United States

Message

What I'm looking for

I seek a role where I can apply research-led engineering to improve model reasoning, fairness, and scalability, collaborating cross-functionally to deploy robust NLP and ML systems.

I am a Machine Learning Engineer and AI Trainer and data annotator with seven years of experience optimizing transformer architectures, improving model fairness, and deploying NLP systems at organizations such as X AI and Facebook AI Research. I lead benchmarking, dataset improvement, and model-debugging efforts that measurably enhance performance and reliability.

My work spans research and engineering: I implement research ideas, build reference implementations, and deploy scalable ML pipelines using TensorFlow, PyTorch, Hugging Face, and cloud platforms. I have published peer-reviewed work on reducing bias in multilingual NLP and developed benchmarks presented at NeurIPS and ACL.

Worked across text, image, audio, and video annotation projects for AI/ML systems

Experience

Work history, roles, and key accomplishments

Current

Machine Learning Engineer

Current

X AI

Jan 2023 - Present (2 years 10 months)

Led optimization and debugging of large transformer models, improving model performance by 20% on key NLP tasks and building benchmarking tasks for summarization, NER, and generation.

Debugging NLP Benchmarking PyTorch TensorFlow Hugging Face

Machine Learning Researcher

Facebook AI Research

Jun 2020 - Dec 2022 (2 years 6 months)

Conducted research on large-scale NLP models improving fairness and interpretability; developed RL algorithms for recommendation systems that increased engagement by 25% and published at NeurIPS and ACL.

NLP FAIR Model Interpretability Reinforcement Learning Research PyTorch Model Evaluation Publishing

Data Scientist

Zindi

Jun 2018 - May 2020 (1 year 11 months)

Developed and deployed ML models and data pipelines for classification, regression, and clustering, improving recommendation accuracy and achieving a 15% uplift in NLP sentiment prediction.

Data Pipelines NLP Python SQL Spark Performance Tuning

Education

Degrees, certifications, and relevant coursework

Harvard University

Master of Science, Computer Science

Activities and societies: Relevant coursework: Advanced Machine Learning, Neural Networks, Natural Language Processing, AI Ethics, Reinforcement Learning.

Master of Science in Computer Science with focus on machine learning, NLP, and deep learning; thesis on optimizing transformer models for multilingual NLP applications.