We want you if you’re:
Actually technical: Prompt engineering is the beginning, not the job.
Reliability-obsessed: “Usually works” bothers you.
Data-minded: Garbage in still means garbage out, even with a very expensive model.
The Why
Anyone can call an LLM API.
We need people who can build AI systems customers can actually trust.
TLDR
Comp: Above-market compensation.
The job: Build reliable production AI systems.
Location: Remote. You can work from anywhere, but you must consistently work Pacific Time. This role sits on our international engineering team.
Culture: High ownership, high intensity.
Structure: Full-time employee and full-time contractor applications welcome.
Experience requirement: Strong engineering fundamentals. “Prompt engineer” alone is not sufficient.
What you’ll do
Build production AI systems.
Design the systems around the model, not merely the prompts sent into it.
Figure out ingestion.
Cleaning.
Normalization.
Retrieval.
Context construction.
Evaluation.
Guardrails.
Fallbacks.
Monitoring.
Reliability.
Work across structured and unstructured customer data.
Build evaluation systems that tell us when AI quality is improving or degrading.
Figure out why something failed instead of simply tweaking the prompt until the demo works.
Your job: Make probabilistic technology reliable enough for real businesses.
We want you if you’re
Systems-minded: The model is one component of the architecture.
Data-minded: You know reliability often begins upstream.
Skeptical: You don't trust an AI system because it worked five times.
Deep engineering + AI fluency + obsession with reliability will win you this job.
Please do NOT apply if you
Need three days to write 300 words.
Think professional writing means removing all personality.
Care more about beautiful sentences than human behavior.
Need perfect briefs before you can start.
If this sounds like you, let’s talk ASAP.
Equal Opportunity: FullCircle is an equal opportunity employer. We consider qualified applicants without regard to any characteristic protected by applicable law.