Darrell S. Best Jr.

Washington, D.C., United States

> DOD TS/SCI Cleared

Read My Articles Contact Me

Professional Summary

  • AI Research Engineer with over a decade of programming experience and 5+ years delivering complex AI/ML solutions.
  • Led AI projects including multilingual LLMs, federated learning systems, FPGA bitstream error correction, and medical text classification.
  • Proven track record in using PyTorch, Hugging Face, DeepSpeed, and Flower to deliver scalable, secure, and efficient AI systems.
  • Developed mission-critical AI solutions for defense, healthcare, and telecommunications, driving operational efficiency and innovation.

Work Experience

Senior Research Engineer, Information Sciences Institute — Arlington, VA

Jul 2019 – Present
  • HealthMap : Mapping unstructured medical reports to ICD codes and predicting conditions with AI.
  • GreenSight : Engineered text encoders with Hugging Face to classify text by Schwartz’s 19 moral values.
  • Sonic Screwdriver : Created FPGA bitstream error correction with custom tokenization, MLM training, and Hugging Face tools.
  • Danube : Built federated learning models with Flower for secure and anomaly-resistant distributed AI training.
  • Hawkeye : Developed multilingual LLMs using GPT-2 and DeepSpeed for scalable foreign language chat systems.

Machine Learning Engineer, QinetiQ US — Lorton, VA

Mar 2018 – Jul 2022
  • HMDS : Built IED detection systems using ground-penetrating radar for Army Husky vehicles and integrated software into Army training simulators.
  • IGSR : Developed border-crossing detection for the FBI using ResNet CNNs for high-accuracy image recognition.
  • Served full-time from March 2018 to July 2019, then part-time consulting from July 2019 to August 2022.

Junior Software Engineer, Windstream — Greenville, SC

Sep 2017 – Mar 2018
  • PUMA: Built end-to-end provisioning software for DSLAMs and network devices using multiple databases and remote connections.

Research Assistant, Clemson University — Clemson, SC

May 2014 – Feb 2018
  • Eye-Tracking Research: Solved the Midas touch problem with natural eye gestures, enabling faster, intuitive user control.
  • Published: A Rotary Dial for Gaze-based PIN Entry (ETRA 2016)

Featured Articles

Multilingual LLM Article

The Future of Multilingual AI

Exploring how multilingual large language models are breaking down language barriers and enabling global communication. This article discusses the challenges and breakthroughs in training models across multiple languages simultaneously.

Research NLP Global AI
Read Article
Federated Learning Article

Privacy-Preserving AI with Federated Learning

How federated learning is revolutionizing data privacy in machine learning. This article examines the architecture, challenges, and real-world applications of training models without centralizing sensitive data.

Privacy Distributed Systems Security
Read Article
Medical AI Article

AI in Healthcare: Transforming Medical Diagnostics

An in-depth look at how AI is revolutionizing medical text classification and diagnosis. This article covers the technical challenges, ethical considerations, and potential future developments in healthcare AI.

Healthcare Ethics Medical AI
Read Article
View All Articles

Education

Publications

A Rotary Dial for Gaze-based PIN Entry

Best, Darrell S. and Duchowski, Andrew T. (2016). In Proceedings of the Ninth Biennial ACM Symposium on Eye Tracking Research & Applications (ETRA ’16), pages 69–76. ACM.

https://doi.org/10.1145/2857491.2857527

Skills

Programming

Python 95%
JavaScript 80%
C++ 75%

AI/ML Frameworks

PyTorch 90%
Hugging Face 85%
TensorFlow 80%

DevOps & Tools

Docker 85%
Git/GitHub 90%
CI/CD 75%

Contact

Email: hireme@darrellbest.com

LinkedIn: https://www.linkedin.com/in/darrellsbest/