SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Tools for Humanity (TFH), the company behind World, is building a planet-scale biometric identity network. World ID has already verified over 17 million people across 160 countries using the Orb, a custom-built iris recognition device. The company combines cutting-edge ML research with hardware deployment to create a trustworthy human verification layer for an AI-driven internet.
You will own the ML lifecycle infrastructure that powers this global system. This role bridges ML research and production deployment, designing and operating reliable pipelines that transform state-of-the-art models into deployed systems running on millions of edge devices (Orbs, Orb Mini, and mobile apps).
Key responsibilities include:
- Design and build production-grade ML infrastructure for training, evaluation, telemetry ingestion, and model deployment
- Maintain and evolve CI/CD workflows and automated pipelines that serve ML research and product teams daily
- Implement edge-aware rollout services with staged deployment, A/B testing, and instant rollback capabilities across distributed hardware
- Develop secure APIs and backend services that expose governed datasets and model artifacts at scale
- Build automated monitoring, drift detection, and alerting systems for real-time model health
- Champion best practices in data lineage, reproducibility, privacy-by-design, and secure edge delivery
- Collaborate with ML research, product, and firmware teams to streamline delivery and feedback loops
You bring 5+ years of experience building ML infrastructure and production ML systems at scale. You have a proven track record delivering platforms and CI/CD pipelines used daily by ML teams. You're hands-on with large-scale GPU cluster training, versioned dataset systems with lineage tracking, containerization (Docker), orchestration (Kubernetes/EKS), and Infrastructure-as-Code (Terraform/CDK/CloudFormation). Strong backend engineering skills in Python and/or Go, deep understanding of modern CI/CD and observability practices, and comfort operating production systems with defined SLAs round out the profile. Experience with agentic AI development is a plus.
