Aug 2026 - Present
software engineer @ Lindy.ai
- building ai employees that take on real business workflows end to end. agents that do the work, not demos that die in a pitch deck.
I co-founded alfred_, an AI executive assistant that handles admin work autonomously. I'm also a Software Engineer at Lindy.ai, building AI employees. I studied Computer Science with a Mathematics minor at the University of Louisiana Monroe.
Most of my time goes into agentic AI. At alfred_, I built a LangGraph pipeline that triages, drafts, and executes email actions on its own, cutting manual inbox work by roughly 80%. At the Simons Foundation, I processed 1.2 TB of BlastoSPIM embryo microscopy data and trained a custom 3D CNN to classify nucleus states. I also collaborate with Dr. Jhim Shim on algorithmic number theory.
I build things that compound. Faster deploys, better datasets, clearer tooling. If you want to talk agentic AI, applied ML, or product engineering. let's connect.
Aug 2026 - Present
2025 - Present
Summer 2025
Aug 2024 - May 2025
2024 - Present
January 2023 - May 2026
October 2022 - May 2023
October 2021 - December 2021
featured
An AI executive assistant that actually does your admin, not just helps you do it faster. Donna, the AI brain behind alfred_, triages your inbox, drafts replies, manages your calendar, and surfaces only what needs your attention. Built with LangGraph multi agent orchestration, unified Gmail/Outlook/Calendar APIs, and a Daily Brief that says "here are the 3 things that need your brain."
An AI platform that joins your Google Meet or Zoom calls, records and transcribes them, then generates structured notes with action items. Organize everything with tags, categories, and search so nothing falls through the cracks.
Plan trips by dropping pins on a map, adding activities at each stop, and letting the app find the fastest route using Dijkstra's algorithm. Built with React and the Google Maps API.
A full stack employee time management system with role based authentication, live GPS clock tracking, and department management. Uses Next.js 14 Server Actions for instant updates and Prisma ORM for data handling, plus automated reports delivered via email.
Simulates pressure distribution around a NACA 0012 airfoil using the finite element method. Built in Python with NumPy for computation and Matplotlib for interactive visualizations.
blogs & tutorials
A pipeline of agents and written rules that takes a story from the open internet to a finished video, and why the interesting part is not the generation but everything that decides what is worth making.
Across three generations of OpenAI models the prices land on the same four numbers. What changes is how much intelligence stands on each rung. That reframes the price war, explains why building on top of frontier models has not got cheaper, and points at where the competition actually moved.
One user's AI-drafted emails kept coming out full of em dashes. The root cause wasn't the model alone, it was our own pipeline teaching itself the habit. Why we stopped asking the model nicely and started enforcing the rule in code.
My inbox is always open, whether you want to talk agentic AI, applied ML, or just say hello. I read everything and reply when I can.