all open roles

research

founding research engineer

build the multimodal agent runtime that lets ciao speak, see, act, remember, and improve in real time.

  • san francisco, ca
  • full-time
  • research

about the role

you’ll build ciao itself: a realtime system that combines voice, browser use, visual perception, product knowledge, and verification. the work starts with real customer sessions and ends when a better behavior is reliable in production.

what you’ll do

  • improve the realtime agent across voice latency, turn-taking, perception, tool use, and action planning.
  • turn product demonstrations into durable skills that work across interface variation.
  • build the knowledge and memory layer that connects product facts, conversation context, and the next useful action.
  • create evaluations and verifiers that explain failures and prove when a fix works.

what we’re looking for

  • deep experience in realtime voice, agents, browser use, multimodal perception, memory systems, or applied ml evaluation.
  • the production engineering ability to carry an experiment into a dependable live system.
  • rigor about what an evaluation proves and curiosity about uncomfortable failure cases.
  • a desire to work directly with product, customers, latency, and reliability constraints.

interested?

tell us who you are and show us the work you’re proud of.

general application

apply to .

tell us who you are and show us the work you’re proud of.

we’ll only use these details to review your application.

Book a demo

Show us your product.

Tell us a little about what you’re building. We’ll follow up personally.

We’ll only use these details to arrange your demo.