Luminal

You'll design and build **core compiler infrastructure** for **AI model optimization**, enabling fast deployment on any hardware. Your work will directly impact production-ready AI across diverse compute platforms. Join a **YC and Felicis-backed startup** as an early engineer.

You'll **deploy and tune models** with optimizations like KV caching and paged attention, **improve the scheduler and batcher** for low-latency serving, and **profile performance and cost** on Luminal Cloud. This founding role offers a chance to shape the inference stack from the ground up.