Skip to content

Details

# Demo Day

Speaker: Eugene Geis, PhD
Title: Director of Research and Data Science
Organization: Bitscopic Inc.

PraediVibe: Building a Trust-Preserving AI

Analyst for Observational Healthcare Data
PraediVibe is an AI-powered analytics assistant designed to generate clinically accurate SQL logic over a very large, anonymized VA healthcare database. Rather than focusing only on natural language to SQL translation, this project emphasizes trust, reproducibility, and guardrails in high-stakes healthcare data environments. In this talk, Eugene will discuss what was built, design constraints, schema constraints, how the agentic workflow works, and some key lessons learned in semantifacturing prototypes of AI systems.

Bio: Dual-PhD data nerd turning messy questions into business strategies for useful products and fewer headaches.

-----

Speaker: Macon Pegram
Title: Developer / Architect
Organization: CapTech Consulting

No Cloud, All Clarity: Crafting a Local Photo Search Engine

Armed with a collection of photographs and a desire to see what's possible with local AI, Macon set off to develop a searchable Photo Catalog. Leaning into local AI first principles and open standards, this project demonstrates a privacy focused, scalable AI architecture. The final product is a web based search tool that allows for multi-modal searching of a photo catalog using local models and vector databases as well as an MCP implementation for chat integration. All of this runs on modest hardware. Macon will share his code and what he learned along the way.
Bio: A technologist who is hype aware but pragmatically driven. When not behind a computer Macon can be found behind a camera or hiking through the woods.

-----

Speaker: Nish Tahir
Title: Principal Applied AI Research Engineer
Organization: Telus Digital

I hate deck building so I trained an AI model to build Magic the Gathering decks for me

Magic the Gathering deck building is a complex task that requires understanding game mechanics, synergies, themes, and combos. Players enjoy hours of fun planning and tweaking their decks. Nish finds it more enjoyable to automate his entertainment instead by training a model to deck build for him.
Bio: Principal Applied AI Research Engineer at Telus Digital.

-----

Speaker: Paul Otto
Title: Software Developer
Organization: Teledyne LeCroy

Using an LLM to Analyze Bluetooth and Wi-Fi Wireless Protocol Data

Analyzing wireless Bluetooth and Wi-Fi protocol captures is hard because of the large data volume and complicated protocol rules. LLMs are great for understanding the rules, but can't digest a complete capture in the prompt. This talk presents an open-source project that lets a LLM analyze the capture in a multi-stage approach without the indigestion.

Project repository:
https://github.com/potto216/wps-examples/tree/main/usecase/llm_rag_ml

Bio: A developer of wireless protocol analyzer software by day and Rust programmer by night.

-----

Speaker: Jill Heinze
Title: AI Trust Strategist & Founder
Organization: Saddle-Stitch Consulting

Building a thinking companion to evaluate risk in AI-enabled software projects

We know that AI can expedite workflows, but how can it also introduce healthy friction into decision-making to help us make systems more humane? That's what I set to find out by building an AI use case evaluation companion that I directed with my thought process, readings, responsible AI standards, and value systems to help me identify opportunities for greater stakeholder involvement, challenging assumptions, and mapping to risk mitigations as I evaluate AI use cases. As a non-engineer, I planned the project and ran tests using Google AI Studio, and have a v1 running locally using Streamlit. I'll talk about what I learned about creating my own companion and about myself in the process.

Bio: Jill Heinze helps product leaders make smarter AI decisions through strategic intelligence and ground truth research. As founder of Saddle-Stitch Consulting, she brings 20 years of user research and competitive intelligence experience to help organizations navigate AI uncertainty, revealing what competitors miss and avoiding expensive mistakes before they happen. She serves as Responsible AI Program Director for The American College of Financial Services and hosts Responsible Tech Talks on LinkedIn Live.

-----

## Event Details

Date: Tuesday, March 24th
Time: 6:00 PM – 8:00 PM
Location: Studio IX – 969 2nd St SE, Charlottesville, VA

***

## Schedule

5:30 PM – Setup begins
5:45 PM — (Optional) Book Club Track
6:00 PM – Doors open (feel free to bring snacks or grab something nearby)
6:30 PM – Demos
7:30 PM – Breakout Track Reports and 3-Minute Pitches
7:45 PM – Open Forum Discussion
Share your projects, demo tools, or bring a topic for group discussion.
8:00 PM – Wrap-up and networking

Organizerss

Owen Zanzal - https://www.linkedin.com/in/owenzanzal/
Mike Powers - https://www.linkedin.com/in/michael-powers-4262033/
Chat with us on Cville Slack #cvilleaiexplorers
If you aren't a member of Cville Slack, DM me your email, and I'll send you an invite.

Sponsors

Studio IX
Coworking space in downtown Charlottesville, Virginia, designed for freelancers, entrepreneurs, and remote workers.
Studio IX offers a flexible shared office environment with options for open workspaces, dedicated desks, and private offices. The space fosters a collaborative community through weekly events and mixers, featuring a 'warm industrial' atmosphere with ample natural light.
Amenities include high-speed internet, soundproof phone booths, conference rooms, and premium coffee. Studio IX is staffed by a community manager on weekdays and provides 24/7 access for members.
Tours of the space are available; interested individuals can contact them via:

Related topics

Events in Charlottesville, VA
AI Algorithms
AI and Society
New Technology
Professional Networking
Technology

You may also like