Company Overview
The company develops infrastructure, tools, and model-access solutions that enable developers and enterprises to build and deploy generative AI applications at scale. Its platform supports generative media workloads through high-performance inference, orchestration, and observability, serving a growing ecosystem of AI-powered products.
Job Description
The Machine Learning Engineer, Safety will own the machine learning models and infrastructure powering the platform’s safety and abuse-detection systems. The role covers the full lifecycle, from developing models that identify harmful content and misuse to building reliable production pipelines that operate at scale.
As part of the Trust and Safety team, the engineer will work closely with safety, security, and infrastructure engineering teams to improve detection accuracy, coverage, latency, and scalability across a platform supporting more than 1,000 models. This hands-on role requires strong machine learning and infrastructure engineering skills, with a focus on enabling rapid AI innovation without compromising safety.
Responsibilities
- Design, develop, maintain, and deploy machine learning models and infrastructure for safety and abuse detection.
- Own safety systems end to end, including model development, inference pipelines, production deployment, and ongoing maintenance.
- Improve the accuracy, coverage, latency, reliability, and scalability of detection pipelines across the platform.
- Collaborate with security and infrastructure engineering teams to integrate safety mechanisms into core platform infrastructure.
- Evaluate and integrate third-party safety tools and vendor models when appropriate.
- Monitor developments in machine learning safety and detection, incorporating relevant techniques and infrastructure patterns into production systems.
- Use large-scale GPU infrastructure to support model inference, evaluation, and experimentation.
- Work with cross-functional teams to deploy new AI capabilities quickly while maintaining effective safety controls.
Qualifications
- Required hands-on experience in trust and safety, content moderation, or abuse-detection systems.
- Strong end-to-end engineering fundamentals, including the ability to own both machine learning models and the infrastructure that serves them in production.
- Ability to independently address ambiguous, high-stakes engineering problems with limited precedent.
- Ability to work on-site in San Francisco five days per week.
Technical Stack
- Programming: Python
- Machine Learning: PyTorch
- Generative AI: Diffusers
- Infrastructure and Orchestration: Kubernetes
- Platform Integration: Python SDK
- Compute: Large-scale GPU infrastructure for inference and evaluation
Work Environment
- The role is based in San Francisco and requires in-person work five days per week.
- The engineer will work alongside dedicated safety engineering, security, and infrastructure teams.
- The position offers opportunities to work with large-scale AI infrastructure and rapidly evolving generative AI technologies.
Disclaimer
SpringCube curates tech job listings from various company websites to support tech professionals globally.
- No Endorsement: Job ads on SpringCube do not imply endorsement of their authenticity or quality.
- No Client Relationship: This company is not a client of SpringCube unless stated.
- To Apply: Click the “Apply” button to be redirected to the hiring company’s application page for this job.
- No Liability: SpringCube is not liable for inaccuracies.