Luke Anglin | AI Engineer

The Mission

My goal is simple: build intelligent systems that generate clear profit—a rarity in today's AI landscape. I focus on deploying Reinforcement Learning and Deep Learning solutions that don't just work, but drive measurable business value through automation and efficiency.

Summary

I specialize in Deep Learning and Distributed Training, optimizing models purely for speed and memory efficiency. Whether architecting cloud-native solutions or fine-tuning on-prem clusters, I ensure infrastructure works as hard as the models it serves.

I bring strong backend software expertise across any environment to help financial institutions maximize measurable savings and profit. I engineer systems that translate technical performance into direct capital efficiency. I am also a Certified CCNA (Cisco Certified Network Associate), providing me with the strong networking fundamentals—routing, switching, and protocol security—essential for robust distributed systems.

University of Virginia

B.S. Computer Science

CCNA Certified

Cisco Systems

MBA Candidate

In Progress

Impact & Projects

Tesla

Adobe

Ernst and Young

UVA Research

Elevance

EMR AI Extraction

Engineered an LLM-based API to extract structured data from medical records using OCR & Computer Vision.

Python LLMs Computer Vision

~$2M Estimated Savings

M&A Agentic RAG

Multimedia Retrieval-Augmented Generation app for M&A teams with a custom recommendation engine.

RAG GenAI Vector DB

200+ Hr Saved Per Project

RL-Based Bid Model

Reinforcement learning model optimizing financial team operations and accuracy.

Reinforcement Learning Analytics

2 Wks Reduction / Year

Enterprise Workflow Automation

Architected LangChain applications that optimized workforce allocation and significantly reduced operational overhead.

LangChain Azure GenAI

~$4M Client Savings

Managed AWS Operations

Orchestrated managed AWS infrastructure to support scalable image and video generation solutions.

AWS Infra-as-Code GenAI Media

2.5M Tokens processed / day

Kafka Parallelism Optimization

Implemented parallelism speedups for high-throughput Kafka operations, enhancing real-time data processing.

Kafka Distributed Systems Big Data

~2x Throughput

Distributed Training Infra

Experimented with novel distributed training techniques across GPU clusters to optimize deep learning model performance.

PyTorch GPU Research

55% Training Speedup (GPU)

IoT Federated Learning

Employed distributed learning on resource-constrained IoT devices using GPS data for public transportation systems.

C++ Federated Learning Edge AI

60% Online vs Offline

Technical Arsenal

Languages

Python C++ Go SQL JavaScript/TypeScript Java

AI & Machine Learning

LLMs RAG Systems PyTorch TensorFlow Reinforcement Learning Computer Vision Distributed Training Diffusion Models

Cloud & DevOps

AWS Azure GCP Docker & Kubernetes Kafka Snowflake CI/CD (Jenkins)

Personal

Service dog training

Surfing

Music

Game engines

Building

Languages

Let's Build

Open to discussing complex systems and new opportunities.