Projects
A selection of my work in machine learning, data science, and software development
Dell Enterprise Hub
As the primary technical lead for Dell Enterprise Hub (DEH) development activities, I led the design and development for Dell Enterprise Hub 2.0, architecting and implementing production LLM inference infrastructure for enterprise customers.
Evaluating LLMs and LLM Systems
I designed and led this technical workshop series to help customer teams tackle one of today's core challenges in AI: how to effectively evaluate large language models (LLMs) when traditional machine learning metrics no longer apply. As LLMs become more accessible, the risks of deploying them without rigorous evaluation have only grown. I created this workshop to equip both engineers and technical leaders with a practical framework for measuring LLM performance, managing risk, and building user trust.
LLM Observability Integration
Initiated and built the integration for the first LLM observability tools on the Hugging Face Hub, leading the planning, integration, documentation, release, and communications for both Arize AI's Phoenix and Langfuse observability platforms.
Essence AI
In this project, I explore the latent representations that exist between lyric-space and pixel-space with generative AI. My goal is to capture the essence of music through visual art.
Semantic Search for Retrieval Augmented Generation (RAG)
I designed and delivered this "Semantic Search for Retrieval-Augmented Generation (RAG)" workshop in August 2023 to give engineering teams a clear, practical roadmap for building search-aware LLM applications. The session opens with the core problem—large language models still "don't know what they don't know." I contrast the cost and rigidity of full model fine-tuning with the agility of RAG, where fresh, domain-specific knowledge is retrieved on demand and injected into the prompt.
Neutralizing Subjectivity Bias with HuggingFace Transformers
The NLP task of text style transfer (TST) aims to automatically control the style attributes of a piece of text while preserving the content, which is an important consideration for making NLP more user-centric.
Automatic Offline Signature Verification
Offline signature verification is a biometric verification task that aims to discriminate between genuine and forged samples of handwritten signatures. This is a particularly important form of verification due to the ubiquitous use of handwritten signatures as a means of personal identification in legal contracts, administrative forms, and financial documents.
Continuous Model Monitoring
After iterations of development and testing, deploying a well-fit machine learning model often feels like the final hurdle for an eager data science team. In practice, however, a trained model is never final, and this milestone marks just the beginning of a new chapter in the ML lifecycle called production ML. This is because most machine learning models are static, but the world we live in is dynamically changing all the time. Changes in environmental conditions like these are referred to as concept drift, and will cause the predictive performance of a model to degrade over time, eventually making it obsolete for the task it was initially intended to solve.
Inferring Concept Drift Without Labeled Data
In practice, a trained machine learning model is never final because the complex relationships that it learns are likely to evolve over time - causing the model's performance to deteriorate if not accounted for.
Object Detection Inference: Visualized
Object detection is a critical task in computer vision - powering use cases such as autonomous driving, surveillance, defect detection in manufacturing, medical image analysis, and more.
Question Answering with BERT
Question answering (QA) is a fundamental NLP task that involves answering questions posed in natural language. This project demonstrates how to build a question answering system using BERT and the SQuAD dataset.
Predicting Phish Setlists with Deep Learning
Built a deep learning model to predict what songs the band, Phish, will play next, treating the problem as a sequential multi-class classification task similar to neural language modeling. The goal was to predict the next song given a prior sequence of songs from the band's extensive catalog of 876 unique songs.