Currently building: Quettos, inference chips for agent workflows

Daniel Molina

Building an AI chip company before 30.

Computer Engineer, University of Waterloo. Founder of Quettos. Building inference chips for agent workflows.

View GitHub Connect on LinkedIn Follow on X Instagram

01 · Vision

The mission

I am building Quettos, an AI chip company, before 30, building inference chips designed for agent workflows.

02 · Focus

What I'm focused on

Inference Chips

Designing silicon purpose-built for AI inference, built around how agents actually run.

Agent Workflows

Studying what agentic workloads demand from hardware: long-horizon reasoning, tool calls, memory. Shaping the chip around it.

Founder Journey

Documenting the path to building an AI chip company before 30, the progress, the lessons, and the silicon along the way.

03 · Work

Starred Projects

RISC-V Pipelined CPU

5-stage pipelined RISC-V (RV32I) processor in Verilog with data forwarding and hazard detection.

RL Logic Optimizer

REINFORCE policy gradient agent for FPGA LUT optimization, minimizing circuit area and depth through learned merge strategies.

04 · Profile

About

Daniel Molina is a Computer Engineering graduate from the University of Waterloo and the founder of Quettos, building inference chips at the intersection of AI, agents, and silicon.

Building something ambitious?
Let's talk.

Get in touch