Monet in Motion: An AI-Powered Interactive Art Exhibit

I built a virtual Claude Monet gallery where your hand controls the experience. Using Claude Code, I created an interactive exhibit that tracks your hand movements through your webcam (powered by MediaPipe Hand Landmarker) to render visual effects and navigate the collection with hand gestures. Beyond art, this same approach has real potential in giving feedback for fitness, physical therapy, and patient support applications.

Check it out →

Curated Tech Event Recommendations in San Francisco

After scrolling through pages of tech events on Luma and emails, I decided to put codex to work. I created an event recommendation tool that recommends events over the next two weeks based on my interests. If you'd like to try it out, check out the link below.

Built with Codex · Designed with Claude · Deployed on Vercel

Check it out →

Voice Agents Have Truly Come to Life in Just 8 Months

In August 2025, I built a multi-language voice agent using ElevenLabs and OpenAI's GPT-5 to automatically match a user's conversation to the right language agent, complete with RAG-powered exam prep and conversation guides for French, Spanish, and Korean.

In April 2026, after a recent conversation with ElevenLabs, I was encouraged to check out their latest improvements. By updating the stack, I can now practice my French with an English-speaking agent who can explain my error and then have a native French speaker say the correct version to hear the native dialect.

Evolution of the stack, 2025 → 2026
LLM: OpenAI GPT-5 → Claude Sonnet 4.6
Voice: ElevenLabs Turbo → Eleven v3, enabling more expressive speech

Statura: A Posture Sensor with Haptic Feedback

In 15.783 Product Design and Development, I worked with a team of mechanical engineers, RISD industrial designers, and fellow Sloan students to build a real working product to solve a problem that a few of us faced sitting too long at desks and slouching watching TV.

Statura is a compact, clip-on posture sensor with haptic feedback and a gamified companion app. The idea came from a simple insight we kept hearing in user interviews: people wanted to improve their posture, but existing solutions were either too obtrusive, too uncomfortable, or too easy to ignore. We spent the semester testing that assumption — going from prototypes taped to t-shirts to a polished alpha prototype with Bluetooth, a 48-hour battery, and an app featuring a little bear that levels up as your posture improves.

As a System Design and Management student, I found engineers optimizing the sensor specs while the design instinct was to make it something you'd actually want to wear. Figuring out how to honor both attributes is an underlying theme that is behind some of the products you love today.

Cold-Climate Autonomy: Winter Operations for Glydways × MIT SDM

In the MIT System Design and Management Core, our team partnered with Glydways, a startup building autonomous personal rapid transit on dedicated tracks, to answer a question they hadn't yet addressed: could their system operate safely in winter?

We evaluated six major architectural decisions and used Monte Carlo simulations to compare options across cost and uptime. Our recommendation was a dedicated autonomous snowplow vehicle with salt distribution. This landed on the Pareto frontier for its low capital cost, minimal infrastructure changes, and a design that fit naturally with Glydways' existing autonomous vehicle ethos.

What stayed with me was how much the problem framing mattered. Winter operability sounds like a technical challenge, but it's really a market expansion enabler. Enabling their architecture to operate in different system conditions allows Glydways to enter more cities!