You have around 1–2+ years of experience and a solid technical foundation. You understand how software systems fit together, how data flows, and why things break. You don't need to be the strongest raw coder in the room, but you can read, reason about, and debug what you and your agents produce
You've actually built and shipped things. Side projects, internships, a first job, hackathon wins, small tools people use. You can point to them and walk us through the choices you made
You've worked in a team. You know how to take feedback, review and get reviewed, and move something forward with other people
You have real product sense for someone early in their career. You can look at a half-formed request and see the job to be done, and you have taste about what makes something good to use
You use AI and agents as your default way of working, and you keep up as the tools change. You treat a model like a fast but unreliable teammate whose work you always check
You move fast and take ownership. You'd rather ship something small and learn than plan something big and wait. You're direct, curious, and happy to challenge how things are done. That fits how we work
A first role or internship at a SaaS or product company
Side projects or small products you built mostly on your own and got into people's hands
Any hands-on experience running evals, watching inference cost, or dealing with model failure modes like hallucination. If you don't have this yet, we'll teach you
This is an early-career role. We care far more about what you've built and how you think than about years on paper