Careers
Member of Technical Staff.
Engineering · Full-time · San Francisco · $180k+
Delphi builds evaluations and training environments for the hardest non-technical questions in AI: how models write, interact, recommend, and exercise judgment. The team is small and the bar is high. We hire people who are truth-seeking, deeply technical, and obsessive about their craft.
About the role
As a Member of Technical Staff, you will build evaluation systems for questions that resist clean answer keys. You will take an ambiguous human judgment, make it measurable, and turn it into a signal models can learn from.
You will work directly with frontier labs, research teams, and domain experts. The systems you ship will set the standard for everything Delphi measures after them.
What you'll do
- Design and ship evaluations for writing, interaction, product judgment, and recommendations.
- Work with frontier labs to expose where capable models still make bad decisions.
- Build systems for adversarial scenarios, subjective tasks, and human evaluation panels.
- Own problems end to end: scoping, prototyping, shipping, and measuring impact.
- Bring signal back from the field and shape what we build next.
You may be a fit if
- You have 3+ years of professional software engineering experience.
- You have strong fundamentals in distributed systems, applied ML, or LLM applications.
- You have shipped and operated production systems on your own.
- You write clearly and have sharp product judgment.
- You hold a high bar and move quickly without being told to.
If this sounds like you, write to us at founders@rundelphi.com with a few sentences about the most impressive thing you have built.