Hi, I’m San.

Second place at the IROS 2026 Robotic Origami Challenge

I entered on my own, as team AxisTilted2. In the final in Pittsburgh, my policy folded the first two stages of a paper plane with two robot hands.

Read the essay

First on both tracks of the NeurIPS 2026 RealPDE competition

I finished the development phase first on both tracks, steering a fleet of AI agents while I travelled to IROS in Pittsburgh. The final phase retrains every finalist from scratch, and results come out on November 10.

Read how

Fifth in the Tartan IMU Challenge at IROS 2026

Chin and I placed fifth of 131 teams. One model had to estimate velocity from one second of raw motion-sensor data, across a car, a legged robot, a drone and a walking person.

The competition

First place at the NeurIPS EAI Challenge

Chin and I won first place in the NeurIPS 2025 Embodied Agent Interface Challenge, as team AxisTilted2. We built language-model agents that reason about physical tasks in simulated environments.

Read the accountResearch paperAward certificate

Joined eBay as an AI researcher

I joined eBay’s Knowledge Extraction team in April 2024, where I research and build language-model systems for information extraction. I was promoted in October 2025.

Current work

Completed my MS at UC San Diego

I completed an MS in computer science at UC San Diego, working with Julian McAuley’s group on AI music and language models and TA’ing recommender systems and data mining.

A UIST award for turning papers into zines

ZINify, which Jaidev Shriram and I built, received an Honorable Mention at the UIST Student Innovation Contest. It uses language models to turn research papers into illustrated zines.

See ZINifyResearch paperAward certificate

A writing-assistant startup at UCSD’s StartR accelerator

I co-founded Glyp, a writing assistant for novelists, and took it through UCSD’s StartR Rady accelerator. It didn’t make it to market, and I wrote about what went wrong.

Read the post-mortem

A summer research internship at eBay

I spent the summer researching information extraction at scale with eBay’s Knowledge Extraction team.

Won the eBay ML Challenge: first of 591 teams

I placed first of 591 teams in the eBay University Machine Learning Challenge, extracting named entities from product titles with DeBERTa-v3 and K-fold ensembling. The result, announced in January 2023, led to the summer research internship.

The announcementAward letter

Moved from chip design to computer science at UCSD

I started the MS at UC San Diego after three years in chip design.

Three years designing chips at Texas Instruments

I spent 2019–2022 on ASIC digital design at Texas Instruments: physical design, RTL, and getting actual chips out the door.

Engineering background

Electrical engineering at NIT Karnataka

I graduated in electrical and electronics engineering. I got into deep learning through Kaggle, earning silver and bronze medals. My thesis on power-quality classification won a best-paper award at IEEE DISCOVER, and the Amateur Astronomy Club gave me a place to explore astronomy.

Best paper certificateKaggle profile