AI/ML Engineer

About the role

ABOUT CINDER

Cinder is the mission-critical infrastructure that keeps the world's most important digital platforms true to what they stand for. The internet has always been abused by bad actors, and AI is making it exponentially worse, driving fraud, abuse, and manipulation at a scale and speed no human team can fight alone. Cinder gives platforms one command center to fight back: to write and enforce policy, deploy AI agents against abuse in real time, investigate threats, file NCMEC reports, and prove their safety programs are working.

Our customers are some of the largest internet platforms in the world. The decisions made in our software directly determine what stays up, what comes down, and how users are treated.

We're a small, fast-moving team backed by Accel and Y Combinator. We care about being intentional, direct, and deeply focused on solving real customer problems.

WHY THIS ROLE

Cinder is expanding quickly, and we want to bring on an additional AI Engineer to partner with our current AI Engineers, Data Scientist, and Data Engineer. You'll build the ML systems that make Cinder faster, more efficient, and more accurate, turning the enormous volume of customer decision data we process into production models that directly shape customer outcomes.

We're looking for a builder. Someone who has taken models from messy data to production at scale, who reaches for a gradient-boosted tree before a transformer when that's the right call, and who has stood up ML infrastructure from scratch rather than inheriting it fully-built. What matters most is judgment: knowing the smallest, most efficient, most reliable model for the job, and understanding both ML methods and LLMs deeply enough to make that call yourself rather than defaulting to whichever one you know best. This role needs someone who cares as much about precision-recall tradeoffs, class imbalance, and serving latency as they do about model architecture.

WHAT YOU'LL DO

- Turn real-world customer data into something a model can actually learn from, then decide what model approach fits: a classical classifier when it wins on cost and latency, a fine-tuned LLM when the tradeoff is worth it, a third-party API as a bootstrap. You own the full path from data to decision, not just the model.

- Improve our classification pipeline, confidence cascading, and detection strategies so we catch harmful content efficiently — balancing cost, latency, and accuracy deliberately.

- Develop intelligent features that help moderators make decisions, organize platform content, and reveal patterns across our data.

- Partner with Engineering to build out Cinder's in-house model training, hosting, and inference platform.

- Design and build the evaluation and metrics infrastructure customers rely on, including how classifier scores and model outputs are calculated, stored, surfaced, and iterated on.

- Partner with our Founding Data Scientist and AI Engineers to shape the agent evaluation architecture — measuring whether our agent fleet is making the right decisions with the right tools at the right cost.