Toggle Light/Dark

Lane Hale

Software Developer → NLP & AI | 27+ Years High-Performance Transaction Systems

PyTorch
Hugging Face
C++
Java
z/TPF Assembler
Email Me lanehale1@gmail.com
Download Résumé (PDF)

About

27-year veteran of airline reservation systems (Sabre, Travelport, Galileo) transitioning to NLP and AI. Recently completed PyTorch and Hugging Face courses (2025) with live projects achieving 97% accuracy on vision models and 85%+ intent accuracy on a flight-assistant chatbot. Passionate about applying Python, C++, Java, and decades of high-performance, mission-critical, application development experience to real-world AI challenges.

Featured Projects

Flight Assistant ChatbotGitHub | Live Demo
BERT-based chatbot with FlightAware + Expedia API integration — 1400+ queries, 85% real-world intent accuracy.

Food Vision MiniGitHub | Live Demo
EfficientNetB2 & ViT-B/16 on Food101 → 97% accuracy.

Geodetic Coordinate ConversionGitHub
C++ implementation of Latitude/Longitude/Altitude to Earth-Centered Earth-Fixed transformation using WGS84 ellipsoid — precision math and modern C++ practices.

Unreal Engine Game Jam (48-Hour Challenge)GitHub
C++ implementation of vector math, randomization, and game logic for a father-son Global Game Jam entry — real-time frame updates and responsive gameplay loop.

Real-Time Video Processing PipelineGitHub
Docker Compose + ZeroMQ + OpenCV — distributed frame processing with multi-stage image optimization (1.66 GB → 808 MB).

Core Skills

Python • PyTorch • Hugging Face • C++ • Java • z/TPF Assembler • Prolog • Linux • Docker • Git • REST APIs • CI/CD • Transaction Systems • NLP • MLOps