Antonio Llano
Hello, I'm Antonio! I'm studying EE & CS at Stanford. Interested in alternative compute, AI4Science, and world models.
[AI for Science] [General] [Hardware]
Previously, I was a 3x IOI & IMO training camp invitee, and once national computing olympiad 1st place winner. I've worked on DNA Computing at Duke University's Reif Group, trained models managing $14MM as a quantitative research intern at Intelneuron, and worked as a software engineer with Sony Pictures and various startups. I also previously started a generative media company funded at $5M valuation from tier-1 VCs and angels.
Recent 10 hour hack →DNA circuits can compute, but most implementations are static logic gates. This project explores sequential decision-making: circuits that observe biomarkers, maintain internal state across multiple measurements, and trigger outputs only after accumulating sufficient evidence. The pipeline compiles POMDP policies into finite-state controllers, maps controller nodes to strand displacement cascades, then uses RL to assign nucleotide sequences that optimize both thermodynamic stability and reaction kinetics. The motivation is enabling more nuanced molecular computation for possible future use cases for in-situ diagnostics and therapeutics.
[report]Exploring whether 2D games can be generated in real-time without traditional engines. Our approach trains action-conditioned diffusion models on gameplay from PPO agents, then aggressively compresses them for consumer hardware through smaller latent spaces, single-step sampling, and simpler attention. Testing if 2D visual simplicity admits efficient architectures that preserve game dynamics under tight latency budgets. Comparing multiple methods to understand quality-compute tradeoffs.
[report]
