Nolan W. Platt

How can we make AI/ML systems more safe, trustworthy, and scalable? How can we reduce the cost of inference in language models while maintaining accuracy and trust? How can we effectively use language models in mission-critical domains?

My name is Nolan Platt, and I seek to answer the above questions rigorously and defensibly through my research. I am a senior at Virginia Tech studying Computer Science.

Currently, I am a Founding Machine Learning Engineer at Oaklet in Berkeley, California, where I am developing an agentic AI ecosystem for a proprietary EHR platform. Previously, I was a Data Science Intern at Hitachi Vantara Federal in Washington D.C., where I focused on machine learning for the federal government.

Current areas of research:

🧠 Domain-Specific LLM Development

First-authored research on automating usability testing in early-stage development. Published at VL/HCC 2025, the premier international confernece on visual languages & human-centric computing.
Worked with Ethan Luchs and Dr. Sehrish Basir Nizamani.

Conferences:
- IEEE VLHCC 2025
- ACM CAPWIC
- CHCI Workshop
- Dennis Dean Conference
- VTURCS Symposium

📊 Natural Language Processing

Exploring quantization methods & synthetic fine-tuning\ for reducing memory costs in small language models. First-authored paper on novel approaches to synthetic fine-tuning and quantization in SLMs. Submitted to IEEE FLLM 2025; pending peer-review.

🔬 Computational Condensed Matter Physics

Worked on the ALPS Library to simulate condensed matter systems.
Mentored by Dr. Fei Lin and Dr. Vito Scarola.
Improving tutorials and contributing to robust numerical simulations used by physicists worldwide.

👨‍🏫 Computer Science Education

Developing infrastructure for OpenDSA, an interactive CS education platform.
Collaborating with Dr. Cliff Shaffer and PhD candidates.
Focusing on backend stability, modernization, and content pipeline upgrades. Led the complete upgrade of the slideshow system used in multiple courses at Virginia Tech; helping advise on course content, assignments, and programming assignments.

All of my research centers around one goal: use computing to better understand our world — I am both fascinated and motivated by the unknown. I focus on interdisciplinary research that aims to make a significant impact on our world.

news

Jul 18, 2025	Paper accepted to IEEE VLHCC 2025!
May 16, 2025	Received President’s List for Spring 2025 for a 4.0 GPA across 22 credit hours
May 14, 2025	Joining Hitachi Vantara as a Data Science Intern!
May 06, 2025	Received the VTCS 2025 Service Award
May 05, 2025	Received the VTCS 2025 David Heilman Research Award
May 02, 2025	Won Best Poster at VT CoE Interdisciplinary Symposium
Apr 22, 2025	Poster accepted to VTURCS Symposium
Apr 22, 2025	Invited to be an Ambassador for the Office of Undergraduate Research for 2025-2026
Apr 14, 2025	Poster accepted to Dennis Dean Conference
Mar 04, 2025	Invited to talk at CHCI Big Idea Workshop
Feb 20, 2025	Short paper accepted to ACM CAPWIC

latest posts

Mar 30, 2025	reflecting on two conferences
Feb 13, 2025	starting off strong

selected publications

IEEE VL/HCC

Catching UX Flaws in Code: Leveraging LLMs to Identify Usability Flaws at the Development Stage

Nolan Platt, E. Luchs, and Sehrish Nizamani

In Proceedings of the 2025 IEEE Symposium on Visual Languages and Human-Centric Computing (VL/HCC), 2025

Bib

@inproceedings{platt2025catching,
  title = {Catching {UX} Flaws in Code: Leveraging {LLMs} to Identify Usability Flaws at the Development Stage},
  author = {Platt, Nolan and Luchs, E. and Nizamani, Sehrish},
  booktitle = {Proceedings of the 2025 IEEE Symposium on Visual Languages and Human-Centric Computing (VL/HCC)},
  pages = {},
  year = {2025},
  month = {},
  publisher = {IEEE,},
}

IEEE FLLM

Multi-Model Synthetic Training for Mission-Critical Small Language Models

Nolan Platt and Pragyansmita Nayak

2025

Submitted to IEEE FLLM 2025 (under review)

Bib

@unpublished{platt2025multimodel,
  title = {Multi-Model Synthetic Training for Mission-Critical Small Language Models},
  author = {Platt, Nolan and Nayak, Pragyansmita},
  note = {Submitted to IEEE FLLM 2025 (under review)},
  year = {2025},
}