ZeroDrift is building the AI-native enforcement infrastructure for enterprise communication, and they are seeking a Principal AI Engineer to own the model that ensures messages are safe to send. This role involves building and maintaining the compliance detection and rewrite model, focusing on accuracy, cost, and latency while working directly with the founder.
Responsibilities:
- Own the compliance detection and rewrite model end to end: data, architecture, training, evaluation, and what runs in production
- Build the evaluation harness and the labeled-data flywheel that prove how good the model is This is your first and highest-leverage priority, because we cannot claim best in the world without it
- Own accuracy, cost, and latency, and the tradeoffs between them
- Own the architecture direction. Today the system combines retrieval, deterministic rules, and LLMs. You decide where it goes next, including whether and when to move to smaller, specialized models
- Build the infrastructure that adapts the model to each customer's rules and policies
- Turn plain-language policy into rules the model can enforce, and own the reasoning behind every decision
- Run tight error analysis on the cases that matter most: violations that should have been caught, and false alarms that erode trust