over 1 year ago
Responsibilities
- Turn language models into useful, functional, and beautiful products.
- Scope and implement novel AI-agent applications, including adversarial webcrawling, weapons identification, red-teaming, and ingredient extraction.
- Architect and implement workflows powering SafetyKit’s risk AI agents.
- Design and conduct experiments evaluating models’ strengths, weaknesses, and optimal use cases.
- Manage the model alpha program with OpenAI and collaborate with its research and product teams.
- Work directly with customers and influence enterprise operations and policy-making.
- Use codegen models to accelerate development.
Requirements
- Strong interest in and knowledge of language models.
- Ability to conduct rigorous evaluations and experiments.
- Ability to learn quickly, take ownership, move quickly, and investigate detailed customer needs.
- Insatiable curiosity and care for the quality and impact of the work.
Categories
About SafetyKit
SafetyKit builds AI agents and a unified risk platform that automates complex trust-and-safety workflows—such as underwriting, investigations, and content moderation—for large marketplaces and payment providers. It sells enterprise SaaS to consolidate rules, models, and human review into a single system for fraud prevention and compliance. The company is privately held, based in San Francisco, and counts Etsy, Patreon, and Discord among its customers.