My Work

A fully deployed, multi-agent AI application where two AI bots debate any topic you throw at them.

Here’s how it works:

You enter any debatable prompt, optionally assign each bot a custom name and role, and set how many turns the debate runs. A moderator bot steps in every two turns to guide the conversation, and at the end, a judge bot scores each participant, provides a summary, and suggests improvements. You can even download a formatted PDF transcript, printable and shareable.

CarbonCoach is a machine learning powered web app that predicts a user’s carbon footprint and provides personalized strategies to reduce it.

This application allows the user to input factors such as travel time, how often they dispose of trash bags, etc. (determined using machine learning models to identify the features that most affect carbon footprint). It then displays their estimated carbon footprint along with a breakdown of which habits contribute most to it.

An LLM, using the Gemini API, provides recommended actionable steps to help users strategically lower their footprint without disrupting their daily routines.

A small open-source tool I built to answer a simple question: what happens when you actually compare AI models side by side instead of trusting whichever one you happened to open?

PlenumView sends a single prompt to multiple LLMs at once and shows the raw responses next to each other, response time, length, everything, no scoring, no "best answer" picked for you. It's MIT licensed, has no build step, and runs locally in under 10 minutes.

It's also the smaller, hands-on proof of the same idea I'm building Plenums Lab around: no single model should get the final word without being weighed against something that thinks differently than it does.

Lockefy is a rewards-based mobile app that turns screen-free time into real-world value. By partnering with local cafés and bookstores, Lockefy incentivizes users to put down their phones at participating locations, earning points for every moment spent offline. Those points can then be redeemed for gift cards at select partner venues, creating a rewarding cycle that benefits both customers and businesses. For partners, Lockefy drives foot traffic and in-store purchases through a performance-based commission model. For users, it transforms mindful moments into meaningful rewards.